TY - JOUR
AU - Devillers, N.
AB - Abstract The objective was to assess sows' lameness by measuring weight distribution on limbs using a force plate made up of 4 individual platforms each resting on 4 single-ended beam load cells. The weight was recorded at an average rate of 14 readings per s over a 15 min period. Ten sows (5 lame sows and 5 sound sows) were weighed twice on 2 different days to assess the repeatability of the measure. Sixty-one sows were then selected in 2 different sites and visually scored for lameness, using a 3-point scoring system (0 = normal gait; 1 = abnormal gait, and/or stiffness; and 2 = shortened stride, and/or the sow puts less weight or avoids putting weight on 1 leg). Various measures were recorded from each sow using the force plate (percentage of weight, the ratio between the weights applied by contralateral legs, weight shifting, and amplitude of weight bearing and weight removing), kinematics (speed, stride length, swing time, stance time, foot height, and carpal and tarsal joints angle average and amplitude), and accelerometers (time spent standing during 24 h, frequency of stepping behavior during feeding, and latency to lie down after feed delivery). The within-sow CV for each measure taken from the force plate were lower than 15%, which reflects a good repeatability. Among force plate measures, only the weight shifting frequency and the ratio between the weights applied by contralateral legs differed between lameness scores. Sows that scored 2 had a higher frequency of weight shifting for fore legs (P = 0.0003) and hind legs (P = 0.0007) than sows scored 0 and 1. The ratio between the weights applied by contralateral legs decreased with the increase of lameness score for the hind limbs (P = 0.014). However, these measures also differed between sites (P < 0.01). These differences may be due to various reasons, including but not limited to genetics and housing systems. Nevertheless, the results suggest that force plate measures such as the asymmetry in the weight applied between a pair of legs and weight shifting are good indicators of lameness. Multivariate analysis on fore and hind legs showed independency between variables related to animals in movement (measures from kinematics) and variables related to static animals (measures from the force plate and accelerometers). Therefore, both static and dynamic methods need to be used to detect various lame sows. INTRODUCTION Lameness is a major concern in animal production and is used as a criterion to assess welfare at the farm level (Veissier et al., 2011). Early detection of lameness is important to provide prompt treatment and hence improve productivity and welfare (Flower et al., 2005). Lameness assessment is typically based on the visual scoring of abnormal gait (Main et al., 2000), abnormal posture (KilBride et al., 2010), or difficulty in lying down (Bonde et al., 2004). Some of these methods have been shown to lack in sensitivity (KilBride et al., 2010) or consistency between observers (D'Eath, 2012). Although visual scoring systems are easy and quick to use at the farm level (Main et al., 2000), there is a need for more objective, reliable, and quantitative assessment systems to improve our understanding of lameness and its causes. Several quantitative methods have been developed to assess gait and lameness. Kinetics and kinematics were used in pigs to evaluate gait in relation to floor surface (von Wachenfelt et al., 2009), and accelerometers have been successfully used in dairy cattle to characterize changes in normal postural patterns, indicating locomotor disorders (Pastell et al., 2009). Kinematics and accelerometers have recently been investigated in sows and were shown to be promising tools to evaluate lameness (Grégoire et al., 2013; Mohling et al., 2014). In dairy cows, the analysis of weight distribution on limbs measured with force plates has been used to detect lame animals (Rushen et al., 2007; Chapinal et al., 2009). This technique was developed in sows but still required further research to evaluate its potential to detect and predict lameness (Sun et al., 2011; Karriker et al., 2013; Pluym et al., 2013; Abell et al., 2014). The objectives of the present study were 1) to validate the measurement of weight distribution of sound sows and sows with naturally occurring lameness with a force plate, 2) to compare the force plate to previously used methods (kinematics and accelerometers), and 3) to identify different types of sows according to their gait and weight distribution patterns. MATERIALS AND METHODS Animals were cared for according to a recommended code of practice (Agriculture and Agri-Food Canada, 1993) and the experimental protocol was approved by the institutional animal care committee (Institutional Animal Care Committee; approval number 390) of the Dairy and Swine R & D Centre (Sherbrooke, QC, Canada) and the University of Saskatchewan's Animal Research Ethics Board, in accordance with the Canadian Council on Animal Care guidelines (CCAC, 2009). Force Plate Design and Procedure The Design. A crate was built by Pacific Industrial Scale Co. Ltd (Richmond, BC, Canada) according to our specifications (Fig. 1). It was sufficiently high and wide (inside dimensions: 213 by 63.5 by 107 cm) to prevent the sows from feeling restrained and becoming agitated. A length adjustment system of 55 cm at the rear of the crate was used for small sows. The force plate was constituted of 4 individual stainless steel platforms (front: 101.6 by 30.5 cm, and rear: 111.8 by 30.5 cm), each resting on 4 single-ended beam load cells. Each platform had a 500 kg capacity (calibrated 250 × 0.1 kg) and was completely independent from the crate. A removable midline bar (203.2 by 1.3 by 15.2 cm) and transversal bar (30.5 by 1.3 by 7.6 cm) were used to ensure that sow's feet remained in their corresponding platform. A feeder with an adjustable feeder gap opening, to ensure a slow distribution of feed, was installed within the crate frame to draw the sow's attention toward a standardized direction and to keep her occupied while standing for 15 min on the force plate. A digital weight indicator (GSE 665) from Pacific Industrial Scale Co. Ltd. simultaneously recorded the BW and weight applied on each separate platform, with an average rate of 14 readings per s. Data were saved using the Pacweight Animal Weight custom software from Pacific Industrial Scale Co. Ltd. Figure 1. View largeDownload slide Description of the force plate, including the crate, a feeder attached to the front door, and 4 independent platforms, each resting on 4 single-ended beam load cells and separated by transversal and horizontal bars. Figure 1. View largeDownload slide Description of the force plate, including the crate, a feeder attached to the front door, and 4 independent platforms, each resting on 4 single-ended beam load cells and separated by transversal and horizontal bars. Two cameras were used to record the position of the sow's legs using the Omnicast video surveillance system (version 4.6, 2001–2012; Genetec Inc., Montreal, QC, Canada), which was synchronized with the Pacweight Animal Weight custom software. The observation started when the sow had her feet placed on the respective platforms and the back adjustment was in place and ended after 15 min had elapsed. Data Cleaning. Video observations were used to detect and remove incorrect periods of recording from weight files (e.g., foot on the wrong platform or sow leaning on the frame). Only periods when the sow stood with her head in the feeder were kept, because they corresponded to a standardized posture. Finally, based on different data cleaning methods used in similar studies on cows (Pastell et al., 2006; Chapinal et al., 2010), any BW per reading higher or lower than 5% of the average BW of the sow was eliminated. Calculations from the Readings. For each reading, the following calculations were made: 1) percentage of weight applied per leg = (weight applied on the leg/BW for the reading) × 100 and 2) ratio between the weights applied by contralateral legs, separately for fore and hind legs = weight applied by the lighter leg divided by the weight applied by the heavier leg (Pluym et al., 2013). Then, for each leg, the average percentage of weight and the SD of the percentage of weight (SD%BW) were calculated. Separately for fore and hind legs, the average ratio between the weights applied by contralateral legs (contralateral ratio) was calculated. Weight shifting (WS) was evaluated by measuring peak characteristics on the weight curves (Fig. 2). Based on calculations used in a sow model (Pluym et al., 2013), a difference of 2.5% BW between the percentage of weight applied on the leg for a reading and the average percentage of weight applied on the leg was considered a peak. Calculations were made per leg: 1) frequency of WS = number of peaks/min, 2) percentage of time WS = (number of readings within peaks/total number of readings) × 100, 3) amplitude of weight bearing = average of all positive peaks amplitude, 4) amplitude of weight removing = average of all negative peaks amplitude, and 5) amplitude of WS = average of the absolute values of positive and negative amplitudes. Figure 2. View largeDownload slide Example of calculation of the different variables measured with the force plate for 1 leg (limits of 2.5% BW). Figure 2. View largeDownload slide Example of calculation of the different variables measured with the force plate for 1 leg (limits of 2.5% BW). Repeatability Study Ten sows (average weight ± SD: 294.4 ± 12.55 kg), 5 visually sound and 5 visually lame sows (either stiff or putting less weight on 1 leg), using the procedure described by Main et al. (2000), were weighed on the force plate apparatus in the morning and afternoon of Day 1 and weighed again 7 d later. Sows were assessed on the force plate using the procedure described previously in the materials and methods. Comparative Study and Characterization of Sows Animals. A total of 61 sows were selected from 2 experimental sites, based on their gait score on a 3-point lameness scoring system adapted from Main et al. (2000; Table 1): 27 Landrace × Yorkshire sows (average weight ± SD: 292.8 ± 21.12 kg) from the Dairy and Swine R&D Centre (DSRDC; Sherbrooke, QC, Canada) housed in individual pens (partially slatted concrete floor) and 34 Landrace × Yorkshire sows (average weight ± SD: 245.3 ± 28.94 kg) from the Prairie Swine Centre Inc. (PSC; Saskatoon, SK, Canada) housed in groups of up to 32 sows in a free access stall system (partially slatted concrete floor, with a rubber mat in some systems). Sows from different parities (n = 1 for parities 0–1 and n = 26 for parities 2–4 at the DSRDC and n = 14 for parities 0–1 and n = 20 for parities 2–8 at the PSC) were between 6 and 10 wk into gestation. In total, 24 sows were scored 0 (n = 11 at the DSRDC and n = 13 at the PSC), 20 sows were scored 1 (n = 8 at the DSRDC and n = 12 at the PSC), and 17 sows were scored 2 (n = 8 at the DSRDC and n = 9 at the PSC). Table 1. Three-point lameness score adapted from Main et al. (2000) and list of criteria used in the detailed visual gait evaluation Item  Description  Lameness score      Score 0  Normal gait, even strides      Score 1  Abnormal gait, stiffness, but lameness not easily identified      Score 2  Lameness detected, shortened strides, sow puts less weight or avoids putting weight on 1 leg  Detailed visual gait evaluation      Avoid bearing weight  Sow lifts her leg to avoid putting weight on it.      Head bob  The head gives jolt when walking      Arched back  The back is arched and the length of the body is shortened.      Asymmetrical stride  Stride's length differs between right and left legs.      Asymmetrical foot placement  Foot placement on the ground (flat, on 1 side of the toe…) differs between right and left legs.      Less weight bearing  Weight distribution differs between right and left. Sow has a slight limp in 1 leg.      Asymmetrical stiffness  Joints look stiff, 1 leg bends less than the opposite 1.      Asymmetrical circular movement  Asymmetrical movement of the legs: 1 leg does outward or inward movement while the other does not.      Asymmetrical swagger of caudal  The posterior sways more on 1 side than the other.  Item  Description  Lameness score      Score 0  Normal gait, even strides      Score 1  Abnormal gait, stiffness, but lameness not easily identified      Score 2  Lameness detected, shortened strides, sow puts less weight or avoids putting weight on 1 leg  Detailed visual gait evaluation      Avoid bearing weight  Sow lifts her leg to avoid putting weight on it.      Head bob  The head gives jolt when walking      Arched back  The back is arched and the length of the body is shortened.      Asymmetrical stride  Stride's length differs between right and left legs.      Asymmetrical foot placement  Foot placement on the ground (flat, on 1 side of the toe…) differs between right and left legs.      Less weight bearing  Weight distribution differs between right and left. Sow has a slight limp in 1 leg.      Asymmetrical stiffness  Joints look stiff, 1 leg bends less than the opposite 1.      Asymmetrical circular movement  Asymmetrical movement of the legs: 1 leg does outward or inward movement while the other does not.      Asymmetrical swagger of caudal  The posterior sways more on 1 side than the other.  View Large Table 1. Three-point lameness score adapted from Main et al. (2000) and list of criteria used in the detailed visual gait evaluation Item  Description  Lameness score      Score 0  Normal gait, even strides      Score 1  Abnormal gait, stiffness, but lameness not easily identified      Score 2  Lameness detected, shortened strides, sow puts less weight or avoids putting weight on 1 leg  Detailed visual gait evaluation      Avoid bearing weight  Sow lifts her leg to avoid putting weight on it.      Head bob  The head gives jolt when walking      Arched back  The back is arched and the length of the body is shortened.      Asymmetrical stride  Stride's length differs between right and left legs.      Asymmetrical foot placement  Foot placement on the ground (flat, on 1 side of the toe…) differs between right and left legs.      Less weight bearing  Weight distribution differs between right and left. Sow has a slight limp in 1 leg.      Asymmetrical stiffness  Joints look stiff, 1 leg bends less than the opposite 1.      Asymmetrical circular movement  Asymmetrical movement of the legs: 1 leg does outward or inward movement while the other does not.      Asymmetrical swagger of caudal  The posterior sways more on 1 side than the other.  Item  Description  Lameness score      Score 0  Normal gait, even strides      Score 1  Abnormal gait, stiffness, but lameness not easily identified      Score 2  Lameness detected, shortened strides, sow puts less weight or avoids putting weight on 1 leg  Detailed visual gait evaluation      Avoid bearing weight  Sow lifts her leg to avoid putting weight on it.      Head bob  The head gives jolt when walking      Arched back  The back is arched and the length of the body is shortened.      Asymmetrical stride  Stride's length differs between right and left legs.      Asymmetrical foot placement  Foot placement on the ground (flat, on 1 side of the toe…) differs between right and left legs.      Less weight bearing  Weight distribution differs between right and left. Sow has a slight limp in 1 leg.      Asymmetrical stiffness  Joints look stiff, 1 leg bends less than the opposite 1.      Asymmetrical circular movement  Asymmetrical movement of the legs: 1 leg does outward or inward movement while the other does not.      Asymmetrical swagger of caudal  The posterior sways more on 1 side than the other.  View Large All sows were assessed using the force plate and, for the purpose of the comparison study, using the accelerometers and kinematics tools, as previously developed (Ringgenberg et al., 2010; Grégoire et al., 2013). Data were collected in the same order for all sows, at an average rate of 5 sows per day. Sows were first fitted with an accelerometer to measure their posture during 24 h (d 0). On Day 2, stepping behavior and force plate measures were taken in the morning, while kinematics and detailed visual gait evaluation were done in the afternoon. Force Plate. The 61 sows were assessed on the force plate using the procedure described previously in the Materials and Methods. Accelerometers. One accelerometer (Hobo Pendant G Data Logger; Onset Computer Corporation, Pocasset, MA), safely protected inside a Velcro pocket and a Vetrap 3M covering (3M Animal Care Products, St. Paul, MN), was placed on 1 of the hind legs of each sow (above the metatarsals) to record standing posture during 24 h, using a method adapted from Ringgenberg et al. (2010). The device recorded the acceleration on the x axis (interval of 5 s) during 24 h. The device was placed with the x axis parallel to the leg, pointing down. When the acceleration data on the x axis was ≥0.59 g, a sow was recorded as standing. To evaluate stepping while standing during feeding time, an accelerometer was placed on each hind leg. The device also recorded the acceleration on the x axis (10 data per s), for 15 min during feeding. A step was considered underway if the x axis acceleration was <0.6 g or >1.4 g (Ringgenberg et al., 2010) while the animal was in a standing position. The latency to lie down after feed delivery was also determined using this recording. Data from recordings were read using the Hoboware Pro software (Onset Computer Corporation). Kinematics. Each sow was video recorded while walking along a corridor (7.3 m long and 1.1 m wide; plain concrete floor type at both experimental sites), using the same procedure as described in Grégoire et al. (2013). Fifteen reflective markers were placed in standardized spots on the sow's body (Fig. 3). Each side of the sow was recorded separately with a DFK22AUC03 camera (The Imaging Source Europe GmbH, Bremen, Germany) with lens (Pentax CCTV C418DX, 4.8 mm, 1:1.8; Pentax Ricoh Imaging Americas Corporation, Denver, CO) using the IC Imaging Capture 2.2 software (The Imaging Source Europe GmbH). Gait characteristics were analyzed using the motion analysis MoviAS Pro software (NAC Image Technology, Simi Valley, CA) for gait characteristics. The data measured were stance time (when the foot was in contact with the floor), swing time (when the foot was in movement), foot height (the foot maximum height during the swing), back angle, back angle amplitude, stride length (distance between sequential floor contacts for a leg), and angle and amplitude of joints angle (carpal joint for fore leg and tarsal joint for hind leg) during the swing and stance periods. The data from 3 individual steps was averaged and this mean was used in the analysis. Figure 3. View largeDownload slide Position of the 15 reflective markers (3 on each leg and 3 on the back) on the sow for the kinematics (back marker is exactly in between the shoulder blades and tail root markers, elbow marker is in the middle of the fore leg sides at the elbow joint level, and stifle marker is in the middle of the hind leg sides at the stifle joint level). Figure 3. View largeDownload slide Position of the 15 reflective markers (3 on each leg and 3 on the back) on the sow for the kinematics (back marker is exactly in between the shoulder blades and tail root markers, elbow marker is in the middle of the fore leg sides at the elbow joint level, and stifle marker is in the middle of the hind leg sides at the stifle joint level). Detailed Visual Gait Evaluation. While sows walked at a steady pace in the kinematics corridor, a more precise and detailed description of the gait was recorded by 1 observer (the same in both sites), where the presence of 9 specific criteria were noted to identify any asymmetrical movement of the body (Table 1), with fore and hind legs being observed separately. Statistical Analysis Repeatability Study. Between-sow and within-sow CV were calculated per leg for the percentage of weight, SD%BW, contralateral ratio, frequency of WS, percentage of time WS, amplitude of weight bearing, and amplitude of weight removing. Between-sow CV (10 sows) were calculated for each of the 4 replicates, and the average CV over the 4 replicates was then calculated per leg. Within-sow CV were calculated between the 4 replicates within each sow, and the average CV over the 10 sows was then calculated per leg. One replicate was missing for 1 sow. Comparative Study. Effect of Lameness Score. Only 60 sows were used for the analyses of the effect of lameness on the measures of each method due to missing or disqualified data. The average value of the right and left sides were calculated to consider values for pair of legs (fore or hind legs). The MIXED procedure of SAS (SAS Inst. Inc., Cary, NC) was used on measures taken by the force plate, with limb position (fore vs. hind legs), visual lameness score (0 vs. 1 vs. 2), and experimental site (DSRDC vs. PSC) in the model. Analyses for the force plate and kinematics measures were then performed using the MIXED procedure of SAS with visual lameness score and experimental site and their interaction in the model, considering fore and hind legs separately. Values are reported as least square mean ± SEM, with the pair of legs as the experimental unit. When residuals indicated a nonconformity to the normality, the Kruskal–Wallis test was performed on the 6 factor combinations (sites × scores) to confirm the results of the ANOVA. Measures of contralateral ratio for fore and hind legs were analyzed using the GLIMMIX procedure of SAS with a β distribution and a logit link function. Values are expressed as back transformed least square mean [with the confidence interval in square brackets]. The stepping behavior (average value for the hind legs) and percentage of time spent standing (accelerometers) were analyzed using the Kruskal–Wallis test on the 6 factor combinations (sites × scores). The sow was the experimental unit. Values are reported as median (lower – upper quartiles in brackets). The latency to lie was categorized as lying within 60 min or not and analyzed using the LOGISTIC procedure of SAS with visual lameness score and experimental site and their interaction in the model. Values are expressed as back transformed least square mean [with the confidence interval in square brackets]. Characterization of Sows. A principal component analysis (PCA) was performed with the PRINCOMP procedure of SAS on all data from kinematics, accelerometers (stepping behavior), and force plate methods to analyze relationships between variables. As per Grégoire et al. (2013), indicating an effect of the position of the limbs (fore vs. hind) on variables, the PCA was performed for fore and hind legs separately. For the PCA, the experimental unit was the leg (right and left). Because the stance time value was significantly correlated to the length of the sow (fore legs: r = 0.61, and hind legs: r = 0.59; P < 0.0001), a correction was applied to this measure by dividing the stance time by the length of the sow. All other measures were not significantly correlated to anatomical measurements of sows. To detect some common patterns among sows, a cluster analysis was done on a selection of variables. Variables were selected according to their potential to discriminate between lameness score and from the PCA. Groups of variables from the PCA were defined according to the distances between variables in relation to their coefficients on the first 5 components. One or two variables were then selected within a group of variables. FASTCLUS and DISCRIM procedures of SAS were used to create categories of legs, regrouping legs with similar values on those selected variables. The decision on the optimal number of categories was based according to the higher value of the pseudo F statistic and cubic clustering criterion (SAS Inst. Inc.). To estimate variables that differed between categories, the MIXED procedure of SAS was used with the category as fixed factor and the leg as repeated measure within a sow in the model. The MIXED procedure of SAS was used with a heterogeneous variance model to assess the difference in contralateral ratio between categories. RESULTS AND DISCUSSION Repeatability Study The average within-sow CV, calculated between 4 repetitions within each of the 10 sows, were lower than 15% for all measurements (Table 2), which can be considered an acceptable within-sow CV (Pluym et al., 2013). Force plate measures can therefore be considered repeatable. Moreover, the within-sow CV were lower than the between-sow CV for each measurement, especially the SD%BW, contralateral ratio, frequency of WS, and percentage of time WS (Table 2). Table 2. Average CV (%) of weight bearing and weight shifting variables calculated between 10 sows over 4 repetitions (between-sow CV) and between 4 repetitions within each sow (within-sow CV)   Between-sow CV  Within-sow CV  Variable1  Fore right  Fore left  Hind right  Hind left  Fore right  Fore left  Hind right  Hind left  Percentage of weight  9.9  10.0  17.6  14.2  6.2  6.0  5.3  4.2  SD%BW  20.4  20.1  32.9  29.0  10.4  11.0  10.9  14.0  Contralateral ratio2  7.4  14.9  3.5  5.4  Frequency of WS  23.9  25.3  24.7  25.8  13.2  12.2  12.6  13.8  Percentage of time WS  8.9  9.8  22.7  20.3  5.3  5.6  9.1  10.4  Amplitude of weight bearing  17.1  21.6  24.5  33.5  12.6  14.2  12.3  13.1  Amplitude of weight removing  19.4  13.4  26.1  19.5  12.5  10.2  11.3  13.5    Between-sow CV  Within-sow CV  Variable1  Fore right  Fore left  Hind right  Hind left  Fore right  Fore left  Hind right  Hind left  Percentage of weight  9.9  10.0  17.6  14.2  6.2  6.0  5.3  4.2  SD%BW  20.4  20.1  32.9  29.0  10.4  11.0  10.9  14.0  Contralateral ratio2  7.4  14.9  3.5  5.4  Frequency of WS  23.9  25.3  24.7  25.8  13.2  12.2  12.6  13.8  Percentage of time WS  8.9  9.8  22.7  20.3  5.3  5.6  9.1  10.4  Amplitude of weight bearing  17.1  21.6  24.5  33.5  12.6  14.2  12.3  13.1  Amplitude of weight removing  19.4  13.4  26.1  19.5  12.5  10.2  11.3  13.5  1SD%BW = SD of the percentage of weight; WS = weight shifting. 2Ratio of the weights applied by contralateral legs. View Large Table 2. Average CV (%) of weight bearing and weight shifting variables calculated between 10 sows over 4 repetitions (between-sow CV) and between 4 repetitions within each sow (within-sow CV)   Between-sow CV  Within-sow CV  Variable1  Fore right  Fore left  Hind right  Hind left  Fore right  Fore left  Hind right  Hind left  Percentage of weight  9.9  10.0  17.6  14.2  6.2  6.0  5.3  4.2  SD%BW  20.4  20.1  32.9  29.0  10.4  11.0  10.9  14.0  Contralateral ratio2  7.4  14.9  3.5  5.4  Frequency of WS  23.9  25.3  24.7  25.8  13.2  12.2  12.6  13.8  Percentage of time WS  8.9  9.8  22.7  20.3  5.3  5.6  9.1  10.4  Amplitude of weight bearing  17.1  21.6  24.5  33.5  12.6  14.2  12.3  13.1  Amplitude of weight removing  19.4  13.4  26.1  19.5  12.5  10.2  11.3  13.5    Between-sow CV  Within-sow CV  Variable1  Fore right  Fore left  Hind right  Hind left  Fore right  Fore left  Hind right  Hind left  Percentage of weight  9.9  10.0  17.6  14.2  6.2  6.0  5.3  4.2  SD%BW  20.4  20.1  32.9  29.0  10.4  11.0  10.9  14.0  Contralateral ratio2  7.4  14.9  3.5  5.4  Frequency of WS  23.9  25.3  24.7  25.8  13.2  12.2  12.6  13.8  Percentage of time WS  8.9  9.8  22.7  20.3  5.3  5.6  9.1  10.4  Amplitude of weight bearing  17.1  21.6  24.5  33.5  12.6  14.2  12.3  13.1  Amplitude of weight removing  19.4  13.4  26.1  19.5  12.5  10.2  11.3  13.5  1SD%BW = SD of the percentage of weight; WS = weight shifting. 2Ratio of the weights applied by contralateral legs. View Large Effect of Limb Position (Fore vs. Hind) and Lameness Score on Force Plate, Kinematics, and Accelerometer Measures Detailed Visual Gait Evaluation. Considering the weight-bearing criteria, only 3.3% of sows put less weight on only 1 fore leg, 3.3% of sows put less weight on fore and hind legs, and 19.7% of sows put less weight on only 1 hind leg. This indicates that more sows were lame on hind legs in this experiment. Force Plate. Fore and hind legs values differed for many variables, independently of the visual lameness score. On average, sows put more weight on a fore leg than on a hind leg (28.8 vs. 21.2% BW, respectively; SEM = 0.14, P < 0.0001). This asymmetry in weight bearing between fore and hind legs has a biomechanical origin, the sow center of gravity being closer to the fore legs (Thorup et al., 2007; Sun et al., 2011). Consequently, a fore leg had a higher SD%BW than a hind leg (7.5 vs. 5.2% BW, respectively; SEM = 0.22, P < 0.0001). Moreover, the amplitude of weight bearing and removing was higher in a fore leg compared to a hind leg (9.0 vs. 7.4% BW, respectively [SEM = 0.19], for amplitude of weight bearing and –5.1 vs. –4.2% BW, respectively [SEM = 0.11], for amplitude of weight removing; P < 0.0001). Fore legs were more often observed WS than hind legs (71.7 vs. 51.3% of time, respectively; SEM = 1.46, P < 0.0001). These results are in accordance with Pluym et al. (2013) who also found a higher frequency of WS in fore legs compared to hind legs in sows. Feeding behavior, due to the slow distribution of concentrate pellets in a feeder in the front door of the scale, may have played a role in the higher frequency of WS in fore legs. Indeed, Pluym et al. (2013) suggested that feeding behavior may increase WS pattern in fore legs. Considering these differences between fore and hind legs, measurements were analyzed separately. Among measures taken from the force plate, only the frequency of WS in fore and hind legs and the contralateral ratio in hind legs differed between visual lameness scores, with no interactive effect with site (P > 0.05; Table 3). Indeed, sows that scored 2 had a significantly higher frequency of WS than sows scored 0 and 1 for fore legs (score 0: 22.5 ± 1.64; score 1: 24.8 ± 1.86; and score 2: 33.3 ± 1.94; P = 0.0003) and hind legs (score 0: 20.4 ± 1.80; score 1: 21.9 ± 2.04; and score 2: 31.3 ± 2.13; P = 0.0007). The contralateral ratio decreased with the increase in lameness score for the hind legs, with sows scored 2 having a lower contralateral ratio than sows scored 0 and 1 (score 0: 0.72 [0.67–0.76]; score 1: 0.71 [0.66–0.75]; and score 2: 0.62 [0.57–0.68]; P = 0.014), with no differences observed for fore legs (P = 0.65). The difference between fore and hind legs in the measure of contralateral ratio was in agreement with the detailed visual gait evaluation indicating that more sows were lame on hind legs. Other measures such as the percentage of weight, the percentage of time WS, and the amplitude of weight bearing and weight removing did not differ between lameness scores (P > 0.05). Because lameness scores were based on sows that were walking as opposed to the force plate measures that were based on a static position, it might be difficult to find relationships between variables that were not measured under the same conditions, as suggested by Pluym et al. (2013). Table 3. Effect of lameness score and site on measures taken with the force plate for fore and hind legs of 60 sows (least square means ± SEM)   Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 11)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 11)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Percentage of weight  28.8 ± 0.32  28.6 ± 0.38  29.2 ± 0.38  28.6 ± 0.30  28.9 ± 0.32  29.0 ± 0.35  0.50  0.85  0.71      SD%BW4  8.75 ± 0.456  7.31 ± 0.534  8.84 ± 0.534  6.20 ± 0.419  7.16 ± 0.456  6.44 ± 0.504  0.72  <0.0001  0.028      Contralateral ratio5  0.63 [0.59–0.66]  0.66 [0.62–0.71]  0.63 [0.58–0.67]  0.73 [0.70–0.76]  0.69 [0.65–0.72]  0.70 [0.66–0.74]  0.65  0.0001  0.12      Frequency of WS,6 per min  26.0 ± 2.41  24.9 ± 2.83  38.4 ± 2.83  19.0 ± 2.22  24.6 ± 2.41  28.1 ± 2.67  0.0003  0.0074  0.17      Percentage of time WS  76.4 ± 2.67  74.5 ± 3.12  78.2 ± 3.12  63.2 ± 2.5  69.9 ± 2.7  68.1 ± 2.9  0.46  0.0002  0.30      Amplitude of weight bearing, % BW  10.3 ± 0.44  8.7 ± 0.51  10.2 ± 0.51  7.9 ± 0.4  9.1 ± 0.4  8.2 ± 0.5  0.87  0.001  0.012      Amplitude of weight removing, % BW  –5.4 ± 0.24  –4.7 ± 0.29  –5.6 ± 0.29  –4.9 ± 0.22  –5.2 ± 0.24  –4.7 ± 0.27  0.69  0.24  0.03  Hind legs3      Percentage of weight  21.2 ± 0.32  21.4 ± 0.38  20.8 ± 0.38  21.4 ± 0.30  21.1 ± 0.32  21.0 ± 0.35  0.50  0.85  0.71      SD%BW  5.71 ± 0.534  4.78 ± 0.626  6.17 ± 0.626  3.95 ± 0.491  5.14 ± 0.534  5.56 ± 0.590  0.16  0.16  0.16      Contralateral ratio  0.67 [0.60–0.73]  0.70 [0.62–0.76]  0.56 [0.48–0.64]  0.77 [0.71–0.82]  0.72 [0.66–0.78]  0.68 [0.60–0.75]  0.014  0.006  0.44      Frequency of WS, per min  23.2 ± 2.64  21.1 ± 3.10  34.2 ± 3.10  17.7 ± 2.43  22.7 ± 2.64  28.1 ± 2.92  0.0007  0.17  0.35      Percentage of time WS  59.8 ± 3.94  47.1 ± 4.62  58.8 ± 4.62  39.2 ± 3.62  48.5 ± 3.94  54.1 ± 4.36  0.12  0.024  0.023      Amplitude of weight bearing, % BW  7.4 ± 0.45  6.8 ± 0.53  7.6 ± 0.53  7.1 ± 0.42  7.8 ± 0.45  7.8 ± 0.50  0.56  0.43  0.44      Amplitude of weight removing, % BW  –4.1 ± 0.28  –3.8 ± 0.33  –4.7 ± 0.33  –3.7 ± 0.26  –4.3 ± 0.28  –4.4 ± 0.31  0.06  0.69  0.28    Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 11)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 11)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Percentage of weight  28.8 ± 0.32  28.6 ± 0.38  29.2 ± 0.38  28.6 ± 0.30  28.9 ± 0.32  29.0 ± 0.35  0.50  0.85  0.71      SD%BW4  8.75 ± 0.456  7.31 ± 0.534  8.84 ± 0.534  6.20 ± 0.419  7.16 ± 0.456  6.44 ± 0.504  0.72  <0.0001  0.028      Contralateral ratio5  0.63 [0.59–0.66]  0.66 [0.62–0.71]  0.63 [0.58–0.67]  0.73 [0.70–0.76]  0.69 [0.65–0.72]  0.70 [0.66–0.74]  0.65  0.0001  0.12      Frequency of WS,6 per min  26.0 ± 2.41  24.9 ± 2.83  38.4 ± 2.83  19.0 ± 2.22  24.6 ± 2.41  28.1 ± 2.67  0.0003  0.0074  0.17      Percentage of time WS  76.4 ± 2.67  74.5 ± 3.12  78.2 ± 3.12  63.2 ± 2.5  69.9 ± 2.7  68.1 ± 2.9  0.46  0.0002  0.30      Amplitude of weight bearing, % BW  10.3 ± 0.44  8.7 ± 0.51  10.2 ± 0.51  7.9 ± 0.4  9.1 ± 0.4  8.2 ± 0.5  0.87  0.001  0.012      Amplitude of weight removing, % BW  –5.4 ± 0.24  –4.7 ± 0.29  –5.6 ± 0.29  –4.9 ± 0.22  –5.2 ± 0.24  –4.7 ± 0.27  0.69  0.24  0.03  Hind legs3      Percentage of weight  21.2 ± 0.32  21.4 ± 0.38  20.8 ± 0.38  21.4 ± 0.30  21.1 ± 0.32  21.0 ± 0.35  0.50  0.85  0.71      SD%BW  5.71 ± 0.534  4.78 ± 0.626  6.17 ± 0.626  3.95 ± 0.491  5.14 ± 0.534  5.56 ± 0.590  0.16  0.16  0.16      Contralateral ratio  0.67 [0.60–0.73]  0.70 [0.62–0.76]  0.56 [0.48–0.64]  0.77 [0.71–0.82]  0.72 [0.66–0.78]  0.68 [0.60–0.75]  0.014  0.006  0.44      Frequency of WS, per min  23.2 ± 2.64  21.1 ± 3.10  34.2 ± 3.10  17.7 ± 2.43  22.7 ± 2.64  28.1 ± 2.92  0.0007  0.17  0.35      Percentage of time WS  59.8 ± 3.94  47.1 ± 4.62  58.8 ± 4.62  39.2 ± 3.62  48.5 ± 3.94  54.1 ± 4.36  0.12  0.024  0.023      Amplitude of weight bearing, % BW  7.4 ± 0.45  6.8 ± 0.53  7.6 ± 0.53  7.1 ± 0.42  7.8 ± 0.45  7.8 ± 0.50  0.56  0.43  0.44      Amplitude of weight removing, % BW  –4.1 ± 0.28  –3.8 ± 0.33  –4.7 ± 0.33  –3.7 ± 0.26  –4.3 ± 0.28  –4.4 ± 0.31  0.06  0.69  0.28  1Number of sows. 2Visual score of lameness (score 0 = normal gait, even strides; score 1 = abnormal gait, stiffness, but lameness not easily identified; and score 2 = lameness detected, shortened strides, sow puts less weight avoids putting weight on 1 leg). 3Average values of right and left legs, for fore legs and hind legs separately. 4SD%BW = SD of the percentage of weight. 5Ratio between the weights applied by contralateral legs (back-transformed mean [with confidence interval in square brackets] are presented). 6WS = weight shifting. View Large Table 3. Effect of lameness score and site on measures taken with the force plate for fore and hind legs of 60 sows (least square means ± SEM)   Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 11)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 11)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Percentage of weight  28.8 ± 0.32  28.6 ± 0.38  29.2 ± 0.38  28.6 ± 0.30  28.9 ± 0.32  29.0 ± 0.35  0.50  0.85  0.71      SD%BW4  8.75 ± 0.456  7.31 ± 0.534  8.84 ± 0.534  6.20 ± 0.419  7.16 ± 0.456  6.44 ± 0.504  0.72  <0.0001  0.028      Contralateral ratio5  0.63 [0.59–0.66]  0.66 [0.62–0.71]  0.63 [0.58–0.67]  0.73 [0.70–0.76]  0.69 [0.65–0.72]  0.70 [0.66–0.74]  0.65  0.0001  0.12      Frequency of WS,6 per min  26.0 ± 2.41  24.9 ± 2.83  38.4 ± 2.83  19.0 ± 2.22  24.6 ± 2.41  28.1 ± 2.67  0.0003  0.0074  0.17      Percentage of time WS  76.4 ± 2.67  74.5 ± 3.12  78.2 ± 3.12  63.2 ± 2.5  69.9 ± 2.7  68.1 ± 2.9  0.46  0.0002  0.30      Amplitude of weight bearing, % BW  10.3 ± 0.44  8.7 ± 0.51  10.2 ± 0.51  7.9 ± 0.4  9.1 ± 0.4  8.2 ± 0.5  0.87  0.001  0.012      Amplitude of weight removing, % BW  –5.4 ± 0.24  –4.7 ± 0.29  –5.6 ± 0.29  –4.9 ± 0.22  –5.2 ± 0.24  –4.7 ± 0.27  0.69  0.24  0.03  Hind legs3      Percentage of weight  21.2 ± 0.32  21.4 ± 0.38  20.8 ± 0.38  21.4 ± 0.30  21.1 ± 0.32  21.0 ± 0.35  0.50  0.85  0.71      SD%BW  5.71 ± 0.534  4.78 ± 0.626  6.17 ± 0.626  3.95 ± 0.491  5.14 ± 0.534  5.56 ± 0.590  0.16  0.16  0.16      Contralateral ratio  0.67 [0.60–0.73]  0.70 [0.62–0.76]  0.56 [0.48–0.64]  0.77 [0.71–0.82]  0.72 [0.66–0.78]  0.68 [0.60–0.75]  0.014  0.006  0.44      Frequency of WS, per min  23.2 ± 2.64  21.1 ± 3.10  34.2 ± 3.10  17.7 ± 2.43  22.7 ± 2.64  28.1 ± 2.92  0.0007  0.17  0.35      Percentage of time WS  59.8 ± 3.94  47.1 ± 4.62  58.8 ± 4.62  39.2 ± 3.62  48.5 ± 3.94  54.1 ± 4.36  0.12  0.024  0.023      Amplitude of weight bearing, % BW  7.4 ± 0.45  6.8 ± 0.53  7.6 ± 0.53  7.1 ± 0.42  7.8 ± 0.45  7.8 ± 0.50  0.56  0.43  0.44      Amplitude of weight removing, % BW  –4.1 ± 0.28  –3.8 ± 0.33  –4.7 ± 0.33  –3.7 ± 0.26  –4.3 ± 0.28  –4.4 ± 0.31  0.06  0.69  0.28    Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 11)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 11)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Percentage of weight  28.8 ± 0.32  28.6 ± 0.38  29.2 ± 0.38  28.6 ± 0.30  28.9 ± 0.32  29.0 ± 0.35  0.50  0.85  0.71      SD%BW4  8.75 ± 0.456  7.31 ± 0.534  8.84 ± 0.534  6.20 ± 0.419  7.16 ± 0.456  6.44 ± 0.504  0.72  <0.0001  0.028      Contralateral ratio5  0.63 [0.59–0.66]  0.66 [0.62–0.71]  0.63 [0.58–0.67]  0.73 [0.70–0.76]  0.69 [0.65–0.72]  0.70 [0.66–0.74]  0.65  0.0001  0.12      Frequency of WS,6 per min  26.0 ± 2.41  24.9 ± 2.83  38.4 ± 2.83  19.0 ± 2.22  24.6 ± 2.41  28.1 ± 2.67  0.0003  0.0074  0.17      Percentage of time WS  76.4 ± 2.67  74.5 ± 3.12  78.2 ± 3.12  63.2 ± 2.5  69.9 ± 2.7  68.1 ± 2.9  0.46  0.0002  0.30      Amplitude of weight bearing, % BW  10.3 ± 0.44  8.7 ± 0.51  10.2 ± 0.51  7.9 ± 0.4  9.1 ± 0.4  8.2 ± 0.5  0.87  0.001  0.012      Amplitude of weight removing, % BW  –5.4 ± 0.24  –4.7 ± 0.29  –5.6 ± 0.29  –4.9 ± 0.22  –5.2 ± 0.24  –4.7 ± 0.27  0.69  0.24  0.03  Hind legs3      Percentage of weight  21.2 ± 0.32  21.4 ± 0.38  20.8 ± 0.38  21.4 ± 0.30  21.1 ± 0.32  21.0 ± 0.35  0.50  0.85  0.71      SD%BW  5.71 ± 0.534  4.78 ± 0.626  6.17 ± 0.626  3.95 ± 0.491  5.14 ± 0.534  5.56 ± 0.590  0.16  0.16  0.16      Contralateral ratio  0.67 [0.60–0.73]  0.70 [0.62–0.76]  0.56 [0.48–0.64]  0.77 [0.71–0.82]  0.72 [0.66–0.78]  0.68 [0.60–0.75]  0.014  0.006  0.44      Frequency of WS, per min  23.2 ± 2.64  21.1 ± 3.10  34.2 ± 3.10  17.7 ± 2.43  22.7 ± 2.64  28.1 ± 2.92  0.0007  0.17  0.35      Percentage of time WS  59.8 ± 3.94  47.1 ± 4.62  58.8 ± 4.62  39.2 ± 3.62  48.5 ± 3.94  54.1 ± 4.36  0.12  0.024  0.023      Amplitude of weight bearing, % BW  7.4 ± 0.45  6.8 ± 0.53  7.6 ± 0.53  7.1 ± 0.42  7.8 ± 0.45  7.8 ± 0.50  0.56  0.43  0.44      Amplitude of weight removing, % BW  –4.1 ± 0.28  –3.8 ± 0.33  –4.7 ± 0.33  –3.7 ± 0.26  –4.3 ± 0.28  –4.4 ± 0.31  0.06  0.69  0.28  1Number of sows. 2Visual score of lameness (score 0 = normal gait, even strides; score 1 = abnormal gait, stiffness, but lameness not easily identified; and score 2 = lameness detected, shortened strides, sow puts less weight avoids putting weight on 1 leg). 3Average values of right and left legs, for fore legs and hind legs separately. 4SD%BW = SD of the percentage of weight. 5Ratio between the weights applied by contralateral legs (back-transformed mean [with confidence interval in square brackets] are presented). 6WS = weight shifting. View Large There was an interaction between site and lameness score for the fore legs SD%BW (P = 0.028; Table 3), amplitude of weight bearing (P = 0.01; Table 3), and amplitude of weight removing (P = 0.03; Table 3). At the DSRDC, sows that scored 1 had lower fore leg SD%BW and lower amplitude of weight bearing and of weight removing than sows scored 0 and 2, while this was not observed at the PSC. There was an interaction between the site and lameness score for the percentage of time WS for hind legs (P = 0.023). Sows that scored 1 at the DSRDC had a lower hind leg percentage of time WS than sows that scored 0 and 2, whereas sows scored 0 at the PSC had a lower percentage of time WS than sows scored 1 and 2. Several measurements were affected by the experimental site independently of the visual lameness score (Table 3). Compared to the PSC, sows at the DSRDC had a lower contralateral ratio for fore legs (0.71 [0.69–0.73]vs. 0.64 [0.62–0.66], respectively; P = 0.0001) and hind legs (0.72 [0.69–0.76] vs. 0.64 [0.60–0.69], respectively; P = 0.006). Compared to the PSC, sows at the DSRDC had a higher frequency of WS for fore legs (23.9 ± 1.41 vs. 29.8 ± 1.56, respectively; P = 0.0074) and a higher percentage of time WS for fore legs (67.0% ± 1.56 vs. 76.3% ± 1.72, respectively; P = 0.0002). Differences observed between sites could be due to many environmental and animal factors, which interact or not with lameness score, for example housing systems, floor type, herd management, parity, or genetics (Pluym et al., 2011). However, because the 2 sites differed on several points in term of rearing and sows characteristics, it was not possible to identify the factors having the most impact. Using a force plate and a transient induced lameness model in sow, Karriker et al. (2013) validated the use of weight distribution (in term of mean weight applied to the legs) as an indicator of lameness. However, in the present study on naturally occurring cases of lameness, the use of the percentage of weight applied by pairs of leg did not help discriminate between lame and sound sows. In cases where the type of lameness and the number of legs affected is unknown, measures such as the asymmetry in the weight applied between contralateral legs and WS appear to be more useful indicators of lameness. This is consistent with force plate studies in cows, reporting that WS was a good indicator of lameness in static animals (Pastell et al., 2010). Karriker et al. (2013) also reported a decrease in contralateral ratio of pressure in lame animals when using a GaitFour Walkway System (CIR Systems, Inc., Havertown, PA) . Kinematics. There were no differences between visual lameness scores for all fore leg measures (P > 0.05; Table 4). Due to the high percentage of sows being lame on their hind legs, differences in kinematics measures between scores were more frequently observed in hind legs, as observed with the force plate measurements. Compared to sows scored 0 and 2, sows that scored 1 were characterized by a lower swing tarsal angle (score 0: 154.4° ± 1.33; score 1: 148.9° ± 1.44; and score 2: 155.9° ± 1.54; P = 0.003), a lower stance tarsal angle (score 0: 150.7° ± 1.36; score 1: 144.8° ± 1.47; and score 2:150.5° ± 1.57; P = 0.008) and a higher amplitude of swing tarsal angle (score 0: 29.4° ± 1.22; score 1: 31.9° ± 1.32; and score 2: 26.3° ± 1.40; P = 0.02). Results indicated an interaction between visual lameness score and site for the measure of stance time for both fore and hind legs (P = 0.007). At the DSRDC, sows that scored 1 had a longer stance time than sows that scored 0 and 2 for fore and hind legs, whereas at the PSC, sows that scored 2 had the longest stance time (Table 4). It clearly appeared that lame sows had a longer stance time at the PSC, similar to Grégoire et al. (2013), but this was not found at the DSRDC. The same observer classified sows according to the 3-points gait score in both sites. Therefore, the difference between sites cannot be explained by a difference in sow classification within lameness score. Because many measures differed between experimental sites, it seems that the gait was different between sows of these 2 herds, regardless of or in addition to their lameness states. Table 4. Effect of lameness score and site on measures of kinematics for fore and hind legs of 60 sows (least square means ± SEM)   Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 10)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 12)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Swing carpal angle,°  177 ± 2.30  174.1 ± 2.58  173.3 ± 2.58  190.7 ± 2.02  185.6 ± 2.10  185.9 ± 2.43  0.12  <0.0001  0.89      Swing carpal angle amplitude, °  60.7 ± 3.62  56.3 ± 4.04  68.2 ± 4.04  47.9 ± 3.17  50.1 ± 3.30  46.7 ± 3.81  0.52  <0.0001  0.14      Stance carpal angle, °  212.0 ± 1.94  208.3 ± 2.17  211.0 ± 2.17  211.6 ± 1.70  208.3 ± 1.77  209.9 ± 2.05  0.19  0.75  0.95      Stance carpal angle amplitude, °  22.0 ± 0.78  22.0 ± 0.87  20.2 ± 0.87  11.1 ± 0.68  12.0 ± 0.71  12.5 ± 0.82  0.71  <0.0001  0.14      Stride length, m  0.875 ± 0.026  0.814 ± 0.029  0.924 ± 0.029  0.819 ± 0.023  0.794 ± 0.024  0.781 ± 0.027  0.14  0.0013  0.08      Foot height, m  0.037 ± 0.003  0.035 ± 0.003  0.035 ± 0.003  0.047 ± 0.003  0.042 ± 0.003  0.043 ± 0.003  0.47  0.003  0.93      Stance time, ms  694.9 ± 28.77  819.3 ± 41.86  659.2 ± 41.86  523.2 ± 32.84  570.8 ± 34.18  664.8 ± 39.47  0.07  <0.0001  0.007      Swing time, ms  448.1 ± 13.39  408.0 ± 14.97  455.7 ± 14.97  425.9 ± 11.74  423.3 ± 12.22  428.5 ± 14.11  0.14  0.31  0.25  Hind legs3      Swing tarsal angle,°  154.2 ± 2.0  150.2 ± 2.24  156.7 ± 2.24  154.7 ± 1.75  147.5 ± 1.83  155.2 ± 2.11  0.003  0.47  0.72      Swing tarsal angle amplitude, °  26.4 ± 1.83  30.0 ± 2.04  26.5 ± 2.04  32.3 ± 1.60  33.8 ± 1.67  26.2 ± 1.92  0.02  0.04  0.24      Stance tarsal angle, °  151.5 ± 2.04  147.8 ± 2.28  152.4 ± 2.28  149.9 ± 1.79  141.8 ± 1.86  148.5 ± 2.15  0.008  0.03  0.54      Stance tarsal angle amplitude, °  13.0 ± 0.95  15.1 ± 1.06  13.0 ± 1.06  11.7 ± 0.83  12.3 ± 0.87  9.9 ± 1.00  0.08  0.004  0.60      Stride length, m  0.872 ± 0.027  0.809 ± 0.031  0.891 ± 0.031  0.807 ± 0.024  0.784 ± 0.025  0.765 ± 0.029  0.28  0.003  0.23      Foot height, m  0.027 ± 0.004  0.035 ± 0.004  0.030 ± 0.004  0.040 ± 0.003  0.041 ± 0.004  0.036 ± 0.004  0.39  0.02  0.61      Stance time, ms  660.6 ± 35.35  791.4 ± 39.52  632.3 ± 39.52  531.7 ± 31.00  571.4 ± 32.27  671.3 ± 37.26  0.05  0.0009  0.004      Swing time, ms  461.6 ± 14.39  438.0 ± 16.08  484.7 ± 16.08  420.9 ± 12.62  420.6 ± 13.13  415.9 ± 15.16  0.38  0.0008  0.25    Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 10)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 12)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Swing carpal angle,°  177 ± 2.30  174.1 ± 2.58  173.3 ± 2.58  190.7 ± 2.02  185.6 ± 2.10  185.9 ± 2.43  0.12  <0.0001  0.89      Swing carpal angle amplitude, °  60.7 ± 3.62  56.3 ± 4.04  68.2 ± 4.04  47.9 ± 3.17  50.1 ± 3.30  46.7 ± 3.81  0.52  <0.0001  0.14      Stance carpal angle, °  212.0 ± 1.94  208.3 ± 2.17  211.0 ± 2.17  211.6 ± 1.70  208.3 ± 1.77  209.9 ± 2.05  0.19  0.75  0.95      Stance carpal angle amplitude, °  22.0 ± 0.78  22.0 ± 0.87  20.2 ± 0.87  11.1 ± 0.68  12.0 ± 0.71  12.5 ± 0.82  0.71  <0.0001  0.14      Stride length, m  0.875 ± 0.026  0.814 ± 0.029  0.924 ± 0.029  0.819 ± 0.023  0.794 ± 0.024  0.781 ± 0.027  0.14  0.0013  0.08      Foot height, m  0.037 ± 0.003  0.035 ± 0.003  0.035 ± 0.003  0.047 ± 0.003  0.042 ± 0.003  0.043 ± 0.003  0.47  0.003  0.93      Stance time, ms  694.9 ± 28.77  819.3 ± 41.86  659.2 ± 41.86  523.2 ± 32.84  570.8 ± 34.18  664.8 ± 39.47  0.07  <0.0001  0.007      Swing time, ms  448.1 ± 13.39  408.0 ± 14.97  455.7 ± 14.97  425.9 ± 11.74  423.3 ± 12.22  428.5 ± 14.11  0.14  0.31  0.25  Hind legs3      Swing tarsal angle,°  154.2 ± 2.0  150.2 ± 2.24  156.7 ± 2.24  154.7 ± 1.75  147.5 ± 1.83  155.2 ± 2.11  0.003  0.47  0.72      Swing tarsal angle amplitude, °  26.4 ± 1.83  30.0 ± 2.04  26.5 ± 2.04  32.3 ± 1.60  33.8 ± 1.67  26.2 ± 1.92  0.02  0.04  0.24      Stance tarsal angle, °  151.5 ± 2.04  147.8 ± 2.28  152.4 ± 2.28  149.9 ± 1.79  141.8 ± 1.86  148.5 ± 2.15  0.008  0.03  0.54      Stance tarsal angle amplitude, °  13.0 ± 0.95  15.1 ± 1.06  13.0 ± 1.06  11.7 ± 0.83  12.3 ± 0.87  9.9 ± 1.00  0.08  0.004  0.60      Stride length, m  0.872 ± 0.027  0.809 ± 0.031  0.891 ± 0.031  0.807 ± 0.024  0.784 ± 0.025  0.765 ± 0.029  0.28  0.003  0.23      Foot height, m  0.027 ± 0.004  0.035 ± 0.004  0.030 ± 0.004  0.040 ± 0.003  0.041 ± 0.004  0.036 ± 0.004  0.39  0.02  0.61      Stance time, ms  660.6 ± 35.35  791.4 ± 39.52  632.3 ± 39.52  531.7 ± 31.00  571.4 ± 32.27  671.3 ± 37.26  0.05  0.0009  0.004      Swing time, ms  461.6 ± 14.39  438.0 ± 16.08  484.7 ± 16.08  420.9 ± 12.62  420.6 ± 13.13  415.9 ± 15.16  0.38  0.0008  0.25  1Number of sows. 2Gait score of lameness (score 0 = normal gait, even strides; score 1 = abnormal gait, stiffness, but lameness not easily identified; and score 2 = lameness detected, shortened strides, sow puts less weight avoids putting weight on 1 leg). 3Average values of right and left legs, for fore legs and hind legs separately. View Large Table 4. Effect of lameness score and site on measures of kinematics for fore and hind legs of 60 sows (least square means ± SEM)   Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 10)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 12)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Swing carpal angle,°  177 ± 2.30  174.1 ± 2.58  173.3 ± 2.58  190.7 ± 2.02  185.6 ± 2.10  185.9 ± 2.43  0.12  <0.0001  0.89      Swing carpal angle amplitude, °  60.7 ± 3.62  56.3 ± 4.04  68.2 ± 4.04  47.9 ± 3.17  50.1 ± 3.30  46.7 ± 3.81  0.52  <0.0001  0.14      Stance carpal angle, °  212.0 ± 1.94  208.3 ± 2.17  211.0 ± 2.17  211.6 ± 1.70  208.3 ± 1.77  209.9 ± 2.05  0.19  0.75  0.95      Stance carpal angle amplitude, °  22.0 ± 0.78  22.0 ± 0.87  20.2 ± 0.87  11.1 ± 0.68  12.0 ± 0.71  12.5 ± 0.82  0.71  <0.0001  0.14      Stride length, m  0.875 ± 0.026  0.814 ± 0.029  0.924 ± 0.029  0.819 ± 0.023  0.794 ± 0.024  0.781 ± 0.027  0.14  0.0013  0.08      Foot height, m  0.037 ± 0.003  0.035 ± 0.003  0.035 ± 0.003  0.047 ± 0.003  0.042 ± 0.003  0.043 ± 0.003  0.47  0.003  0.93      Stance time, ms  694.9 ± 28.77  819.3 ± 41.86  659.2 ± 41.86  523.2 ± 32.84  570.8 ± 34.18  664.8 ± 39.47  0.07  <0.0001  0.007      Swing time, ms  448.1 ± 13.39  408.0 ± 14.97  455.7 ± 14.97  425.9 ± 11.74  423.3 ± 12.22  428.5 ± 14.11  0.14  0.31  0.25  Hind legs3      Swing tarsal angle,°  154.2 ± 2.0  150.2 ± 2.24  156.7 ± 2.24  154.7 ± 1.75  147.5 ± 1.83  155.2 ± 2.11  0.003  0.47  0.72      Swing tarsal angle amplitude, °  26.4 ± 1.83  30.0 ± 2.04  26.5 ± 2.04  32.3 ± 1.60  33.8 ± 1.67  26.2 ± 1.92  0.02  0.04  0.24      Stance tarsal angle, °  151.5 ± 2.04  147.8 ± 2.28  152.4 ± 2.28  149.9 ± 1.79  141.8 ± 1.86  148.5 ± 2.15  0.008  0.03  0.54      Stance tarsal angle amplitude, °  13.0 ± 0.95  15.1 ± 1.06  13.0 ± 1.06  11.7 ± 0.83  12.3 ± 0.87  9.9 ± 1.00  0.08  0.004  0.60      Stride length, m  0.872 ± 0.027  0.809 ± 0.031  0.891 ± 0.031  0.807 ± 0.024  0.784 ± 0.025  0.765 ± 0.029  0.28  0.003  0.23      Foot height, m  0.027 ± 0.004  0.035 ± 0.004  0.030 ± 0.004  0.040 ± 0.003  0.041 ± 0.004  0.036 ± 0.004  0.39  0.02  0.61      Stance time, ms  660.6 ± 35.35  791.4 ± 39.52  632.3 ± 39.52  531.7 ± 31.00  571.4 ± 32.27  671.3 ± 37.26  0.05  0.0009  0.004      Swing time, ms  461.6 ± 14.39  438.0 ± 16.08  484.7 ± 16.08  420.9 ± 12.62  420.6 ± 13.13  415.9 ± 15.16  0.38  0.0008  0.25    Dairy and Swine R&D Centre  Prairie Swine Centre Inc.  Effect  Variable  Score 0 (n = 10)1  Score 1 (n = 8)  Score 2 (n = 8)  Score 0 (n = 13)  Score 1 (n = 12)  Score 2 (n = 9)  Score2  Site  Interaction  Fore legs3      Swing carpal angle,°  177 ± 2.30  174.1 ± 2.58  173.3 ± 2.58  190.7 ± 2.02  185.6 ± 2.10  185.9 ± 2.43  0.12  <0.0001  0.89      Swing carpal angle amplitude, °  60.7 ± 3.62  56.3 ± 4.04  68.2 ± 4.04  47.9 ± 3.17  50.1 ± 3.30  46.7 ± 3.81  0.52  <0.0001  0.14      Stance carpal angle, °  212.0 ± 1.94  208.3 ± 2.17  211.0 ± 2.17  211.6 ± 1.70  208.3 ± 1.77  209.9 ± 2.05  0.19  0.75  0.95      Stance carpal angle amplitude, °  22.0 ± 0.78  22.0 ± 0.87  20.2 ± 0.87  11.1 ± 0.68  12.0 ± 0.71  12.5 ± 0.82  0.71  <0.0001  0.14      Stride length, m  0.875 ± 0.026  0.814 ± 0.029  0.924 ± 0.029  0.819 ± 0.023  0.794 ± 0.024  0.781 ± 0.027  0.14  0.0013  0.08      Foot height, m  0.037 ± 0.003  0.035 ± 0.003  0.035 ± 0.003  0.047 ± 0.003  0.042 ± 0.003  0.043 ± 0.003  0.47  0.003  0.93      Stance time, ms  694.9 ± 28.77  819.3 ± 41.86  659.2 ± 41.86  523.2 ± 32.84  570.8 ± 34.18  664.8 ± 39.47  0.07  <0.0001  0.007      Swing time, ms  448.1 ± 13.39  408.0 ± 14.97  455.7 ± 14.97  425.9 ± 11.74  423.3 ± 12.22  428.5 ± 14.11  0.14  0.31  0.25  Hind legs3      Swing tarsal angle,°  154.2 ± 2.0  150.2 ± 2.24  156.7 ± 2.24  154.7 ± 1.75  147.5 ± 1.83  155.2 ± 2.11  0.003  0.47  0.72      Swing tarsal angle amplitude, °  26.4 ± 1.83  30.0 ± 2.04  26.5 ± 2.04  32.3 ± 1.60  33.8 ± 1.67  26.2 ± 1.92  0.02  0.04  0.24      Stance tarsal angle, °  151.5 ± 2.04  147.8 ± 2.28  152.4 ± 2.28  149.9 ± 1.79  141.8 ± 1.86  148.5 ± 2.15  0.008  0.03  0.54      Stance tarsal angle amplitude, °  13.0 ± 0.95  15.1 ± 1.06  13.0 ± 1.06  11.7 ± 0.83  12.3 ± 0.87  9.9 ± 1.00  0.08  0.004  0.60      Stride length, m  0.872 ± 0.027  0.809 ± 0.031  0.891 ± 0.031  0.807 ± 0.024  0.784 ± 0.025  0.765 ± 0.029  0.28  0.003  0.23      Foot height, m  0.027 ± 0.004  0.035 ± 0.004  0.030 ± 0.004  0.040 ± 0.003  0.041 ± 0.004  0.036 ± 0.004  0.39  0.02  0.61      Stance time, ms  660.6 ± 35.35  791.4 ± 39.52  632.3 ± 39.52  531.7 ± 31.00  571.4 ± 32.27  671.3 ± 37.26  0.05  0.0009  0.004      Swing time, ms  461.6 ± 14.39  438.0 ± 16.08  484.7 ± 16.08  420.9 ± 12.62  420.6 ± 13.13  415.9 ± 15.16  0.38  0.0008  0.25  1Number of sows. 2Gait score of lameness (score 0 = normal gait, even strides; score 1 = abnormal gait, stiffness, but lameness not easily identified; and score 2 = lameness detected, shortened strides, sow puts less weight avoids putting weight on 1 leg). 3Average values of right and left legs, for fore legs and hind legs separately. View Large To conclude, kinematics appeared to give different results mostly for tarsal joint measures of sows that scored 1. This confirms again that lameness affects mostly hind limbs. The fact that only sows that scored 1 were different from the others could be due to the criteria used for visual scoring. Indeed, score 1 reflects something unclear (a stiffness) in the gait that seemed to be highlighted by the kinematics through the measure of joint angles, whereas score 2 relies more on the avoidance of weight bearing, which was detected by the force plate. Accelerometers. Although there were no significant differences between lameness scores in the percentage of time spent standing during a day (score 0: 17.4%; score 1: 16.3% (6.5-22.8); and score 2: 10.9% (8.4–17.4); P = 0.30), the results are numerically in accordance with the findings from Grégoire et al. (2013), showing a decrease of time spent standing with lameness. For the latency to lie down after feed delivery, the probability to lie down was 0.44 [0.25-0.65] for sows that scored 0, 0.52 [0.30–0.74] for sows that scored 1, and 0.79 [0.51–0.93] for sows that scored 2 (P = 0.14). The percentage of sows that lie down in the first hour following feed delivery was therefore numerically higher for lame sows. There were significant differences between groups in the stepping measures (2.1 steps/min (1.8–3.1) for score 0, 3.2 steps/min (2.0–4.1) for score 1, and 6.9 steps/min (4.1–8.3( for score 2 at the DSRDC and 3.8 steps/min (3.3–4.7) for score 0, 7.0 steps/min (4.8–9.7) for score 1, and 7.1 steps/min (4.8–14.6) for score 2 at the PSC; P < 0.0001). An increase of stepping behavior with lameness is in accordance with findings from Grégoire et al. (2013). A difference between sites was also observed for the other measures from the force plate and kinematics methods. This confirmed the importance of considering the site characteristics (housing system and sow characteristics) when interpreting measures of lameness (Conte et al., 2014). Nevertheless, these results confirm the usefulness of accelerometers in the identification of lameness through stepping behavior when the animal is standing. Results also highlighted that recording for 15 min after feed delivery was enough to detect differences in stepping behavior between lameness scores. Comparison of the Different Lameness Assessment Methods Fore Legs. The eigenvalues of the correlation matrix were 4.35 on principal component (Prin) 1 (29.02%), 2.67 on Prin2 (17.77%), 1.45 on Prin3 (9.64%), 1.15 on Prin4 (7.64%), and 1.05 on Prin5 (6.97%). Four groups of variables can be seen from the first 5 components (Table 5). The first group was mainly composed of weight distribution variables as measured with the force plate: percentage of time WS, SD%BW, and amplitude of WS and to a lesser extent the frequency of WS. Measures of swing time and stride length also appeared to be linked together and to the swing carpal angle amplitude. The third group included measures of foot height, stance time, and back angle. Stance time appeared to be negatively linked to foot height and back angle. Finally, stance carpal angle amplitude and back angle amplitude were close to each other and characterized group 4. Table 5. Eigenvectors values from the principal component analysis for fore and hind legs (the experimental unit is the leg)   Fore legs (n = 118)2    Hind legs (n = 117)  Variable1  Prin13 (29.02%)4  Prin2 (17.77%)  Prin3 (9.64%)  Prin4 (7.64%)  Prin5 (6.97%)  Group  Prin1 (22.67%)  Prin2 (14.69%)  Prin3 (12.81%)  Prin4 (10.11%)  Prin5 (7.33%)  Group  Percentage of weight  0.04  –0.01  0.19  –0.34  0.76    –0.11  –0.03  0.28  0.38  –0.06    SD%BW  0.39  0.18  0.16  –0.32  –0.20  1  0.48  0.06  –0.08  0.14  –0.03  1  Percentage of time WS  0.39  0.13  0.18  –0.17  –0.17  1  0.47  0.07  0.03  0.13  –0.04  1  Frequency of WS  0.28  0.01  0.21  0.21  –0.01  1  0.38  0.11  0.11  –0.08  0.14  1  Amplitude of WS  0.33  0.24  0.13  –0.36  –0.13  1  0.43  0.02  –0.22  0.14  0.09  1  Stride length  0.30  0.21  –0.42  0.06  0.07  2  0.16  –0.28  0.29  0.27  –0.08  2  Swing time  0.20  0.27  –0.21  0.37  0.04  2  0.23  –0.10  0.46  0.12  –0.04  2  Stance time  0.16  –0.43  0.22  0.10  0.01  3  0.04  0.10  0.34  –0.44  –0.23  5  Foot height  –0.06  0.39  –0.25  –0.08  0.28  3  –0.04  –0.24  0.11  0.15  0.45  4  Swing angle5  –0.22  0.39  0.44  0.26  0.01    –0.12  0.55  0.11  0.23  0.23  3  Stance angle  0.13  0.29  0.43  0.35  0.21    –0.10  0.51  0.18  0.28  0.27  3  Swing angle amplitude  0.32  –0.03  –0.34  0.18  0.31  2  –0.05  –0.46  0.05  0.01  0.46  4  Stance angle Amplitude  0.33  –0.29  0.02  –0.01  –0.02  4  –0.04  –0.07  0.26  –0.18  0.43  4  Back angle  –0.10  0.29  –0.16  0.02  –0.34  3  0.00  –0.16  –0.11  0.43  –0.24    Back angle amplitude  0.24  –0.20  0.02  0.45  –0.01  4  0.12  0.04  0.49  –0.25  –0.11  5  Step  –  –  –  –  –    0.28  0.10  –0.27  –0.29  0.33      Fore legs (n = 118)2    Hind legs (n = 117)  Variable1  Prin13 (29.02%)4  Prin2 (17.77%)  Prin3 (9.64%)  Prin4 (7.64%)  Prin5 (6.97%)  Group  Prin1 (22.67%)  Prin2 (14.69%)  Prin3 (12.81%)  Prin4 (10.11%)  Prin5 (7.33%)  Group  Percentage of weight  0.04  –0.01  0.19  –0.34  0.76    –0.11  –0.03  0.28  0.38  –0.06    SD%BW  0.39  0.18  0.16  –0.32  –0.20  1  0.48  0.06  –0.08  0.14  –0.03  1  Percentage of time WS  0.39  0.13  0.18  –0.17  –0.17  1  0.47  0.07  0.03  0.13  –0.04  1  Frequency of WS  0.28  0.01  0.21  0.21  –0.01  1  0.38  0.11  0.11  –0.08  0.14  1  Amplitude of WS  0.33  0.24  0.13  –0.36  –0.13  1  0.43  0.02  –0.22  0.14  0.09  1  Stride length  0.30  0.21  –0.42  0.06  0.07  2  0.16  –0.28  0.29  0.27  –0.08  2  Swing time  0.20  0.27  –0.21  0.37  0.04  2  0.23  –0.10  0.46  0.12  –0.04  2  Stance time  0.16  –0.43  0.22  0.10  0.01  3  0.04  0.10  0.34  –0.44  –0.23  5  Foot height  –0.06  0.39  –0.25  –0.08  0.28  3  –0.04  –0.24  0.11  0.15  0.45  4  Swing angle5  –0.22  0.39  0.44  0.26  0.01    –0.12  0.55  0.11  0.23  0.23  3  Stance angle  0.13  0.29  0.43  0.35  0.21    –0.10  0.51  0.18  0.28  0.27  3  Swing angle amplitude  0.32  –0.03  –0.34  0.18  0.31  2  –0.05  –0.46  0.05  0.01  0.46  4  Stance angle Amplitude  0.33  –0.29  0.02  –0.01  –0.02  4  –0.04  –0.07  0.26  –0.18  0.43  4  Back angle  –0.10  0.29  –0.16  0.02  –0.34  3  0.00  –0.16  –0.11  0.43  –0.24    Back angle amplitude  0.24  –0.20  0.02  0.45  –0.01  4  0.12  0.04  0.49  –0.25  –0.11  5  Step  –  –  –  –  –    0.28  0.10  –0.27  –0.29  0.33    1SD%BW = SD of the percentage of weight; WS = weight shifting. 2Number of legs. 3Prin = principal component. 4Percent of the total variation. 5Carpal angle for fore legs and tarsal angle for hind legs. View Large Table 5. Eigenvectors values from the principal component analysis for fore and hind legs (the experimental unit is the leg)   Fore legs (n = 118)2    Hind legs (n = 117)  Variable1  Prin13 (29.02%)4  Prin2 (17.77%)  Prin3 (9.64%)  Prin4 (7.64%)  Prin5 (6.97%)  Group  Prin1 (22.67%)  Prin2 (14.69%)  Prin3 (12.81%)  Prin4 (10.11%)  Prin5 (7.33%)  Group  Percentage of weight  0.04  –0.01  0.19  –0.34  0.76    –0.11  –0.03  0.28  0.38  –0.06    SD%BW  0.39  0.18  0.16  –0.32  –0.20  1  0.48  0.06  –0.08  0.14  –0.03  1  Percentage of time WS  0.39  0.13  0.18  –0.17  –0.17  1  0.47  0.07  0.03  0.13  –0.04  1  Frequency of WS  0.28  0.01  0.21  0.21  –0.01  1  0.38  0.11  0.11  –0.08  0.14  1  Amplitude of WS  0.33  0.24  0.13  –0.36  –0.13  1  0.43  0.02  –0.22  0.14  0.09  1  Stride length  0.30  0.21  –0.42  0.06  0.07  2  0.16  –0.28  0.29  0.27  –0.08  2  Swing time  0.20  0.27  –0.21  0.37  0.04  2  0.23  –0.10  0.46  0.12  –0.04  2  Stance time  0.16  –0.43  0.22  0.10  0.01  3  0.04  0.10  0.34  –0.44  –0.23  5  Foot height  –0.06  0.39  –0.25  –0.08  0.28  3  –0.04  –0.24  0.11  0.15  0.45  4  Swing angle5  –0.22  0.39  0.44  0.26  0.01    –0.12  0.55  0.11  0.23  0.23  3  Stance angle  0.13  0.29  0.43  0.35  0.21    –0.10  0.51  0.18  0.28  0.27  3  Swing angle amplitude  0.32  –0.03  –0.34  0.18  0.31  2  –0.05  –0.46  0.05  0.01  0.46  4  Stance angle Amplitude  0.33  –0.29  0.02  –0.01  –0.02  4  –0.04  –0.07  0.26  –0.18  0.43  4  Back angle  –0.10  0.29  –0.16  0.02  –0.34  3  0.00  –0.16  –0.11  0.43  –0.24    Back angle amplitude  0.24  –0.20  0.02  0.45  –0.01  4  0.12  0.04  0.49  –0.25  –0.11  5  Step  –  –  –  –  –    0.28  0.10  –0.27  –0.29  0.33      Fore legs (n = 118)2    Hind legs (n = 117)  Variable1  Prin13 (29.02%)4  Prin2 (17.77%)  Prin3 (9.64%)  Prin4 (7.64%)  Prin5 (6.97%)  Group  Prin1 (22.67%)  Prin2 (14.69%)  Prin3 (12.81%)  Prin4 (10.11%)  Prin5 (7.33%)  Group  Percentage of weight  0.04  –0.01  0.19  –0.34  0.76    –0.11  –0.03  0.28  0.38  –0.06    SD%BW  0.39  0.18  0.16  –0.32  –0.20  1  0.48  0.06  –0.08  0.14  –0.03  1  Percentage of time WS  0.39  0.13  0.18  –0.17  –0.17  1  0.47  0.07  0.03  0.13  –0.04  1  Frequency of WS  0.28  0.01  0.21  0.21  –0.01  1  0.38  0.11  0.11  –0.08  0.14  1  Amplitude of WS  0.33  0.24  0.13  –0.36  –0.13  1  0.43  0.02  –0.22  0.14  0.09  1  Stride length  0.30  0.21  –0.42  0.06  0.07  2  0.16  –0.28  0.29  0.27  –0.08  2  Swing time  0.20  0.27  –0.21  0.37  0.04  2  0.23  –0.10  0.46  0.12  –0.04  2  Stance time  0.16  –0.43  0.22  0.10  0.01  3  0.04  0.10  0.34  –0.44  –0.23  5  Foot height  –0.06  0.39  –0.25  –0.08  0.28  3  –0.04  –0.24  0.11  0.15  0.45  4  Swing angle5  –0.22  0.39  0.44  0.26  0.01    –0.12  0.55  0.11  0.23  0.23  3  Stance angle  0.13  0.29  0.43  0.35  0.21    –0.10  0.51  0.18  0.28  0.27  3  Swing angle amplitude  0.32  –0.03  –0.34  0.18  0.31  2  –0.05  –0.46  0.05  0.01  0.46  4  Stance angle Amplitude  0.33  –0.29  0.02  –0.01  –0.02  4  –0.04  –0.07  0.26  –0.18  0.43  4  Back angle  –0.10  0.29  –0.16  0.02  –0.34  3  0.00  –0.16  –0.11  0.43  –0.24    Back angle amplitude  0.24  –0.20  0.02  0.45  –0.01  4  0.12  0.04  0.49  –0.25  –0.11  5  Step  –  –  –  –  –    0.28  0.10  –0.27  –0.29  0.33    1SD%BW = SD of the percentage of weight; WS = weight shifting. 2Number of legs. 3Prin = principal component. 4Percent of the total variation. 5Carpal angle for fore legs and tarsal angle for hind legs. View Large Hind Legs. The eigenvalues of the correlation matrix were 3.63 on Prin1 (22.67%), 2.35 on Prin2 (14.69%), 2.05 on Prin3 (12.81%), 1.62 on Prin4 (10.11%), and 1.17 on Prin5 (7.33%). Five groups of variables can be seen from the first 3 Prin (Table 5). The first group was composed of weight distribution variables: frequency of WS, percentage of time WS, SD%BW, and amplitude of WS. The second group was composed of stride length and swing time. A third group was composed of measures of tarsal angle during the swing and the stance phases. The fourth group was composed of measures of foot height and swing and stance tarsal angle amplitudes. A final group comprised stance time and back angle amplitude. In summary, multivariate analysis on fore and hind legs tended to show independency between variables related to animals in movement (carpal or tarsal swing angles and swing and stance times) and variables related to static animals (WS and stepping). This suggests that the 2 methods should be used when assessing lameness. This independence may be related to the different causes of lameness and the number of limbs affected. For instance, pain in 1 limb will be expressed by WS while pain in all feet will be expressed by a reluctance to move (Wells, 1984). Moreover, some types of lameness, such as those associated with joint problems, may be visually observed but go undetected with a force plate (Pluym et al., 2013). Therefore, combined observation of static posture and movement patterns has been suggested in studies on cows (Pastell et al., 2010). Characterization of Sows According to their Lameness Pattern A categorization process was performed on fore and hind legs separately according to different variables selected to best reflect characteristics of sow legs. Five variables in common for fore and hind legs were selected: percentage of time WS, stance time, swing angle (carpal and tarsal), foot height, and stride length. A measure of stance carpal angle was added for fore legs because of its relative independency from other variables in the PCA, and stepping behavior was added for the hind legs because it is a good indicator of lameness. The optimal number of categories was 3 for fore legs (pseudo F statistic = 29.61 and cubic clustering criterion = 7.6). The 3 categories for fore legs are F1, F2 and F3. The optimal number of categories was 3 for hind legs (pseudo F statistic = 18.87 and cubic clustering criterion = 0.09). However, because of the presence of sows having an extreme value for stepping, a fourth category was considered for hind legs to include these extreme sows. The 4 categories for hind legs are H1, H2, H3, and H4. Table 6 shows the mean values of variables used in the cluster analysis for each category of fore and hind legs. Table 6. Categorization of fore legs into 3 groups (F1, F2 and F3) and hind legs into 4 groups (H1, H2, H3 and H4) according to cluster analysis using 6 variables from kinematics, force plate, and accelerometer methods. Values are presented as least square means ± SEM (minimum–maximum). The leg is the experimental unit.   Fore legs  Hind legs  Variable  F1 (n = 14)1  F2 (n = 65)  F3 (n = 39)  H1 (n = 33)  H2 (n = 50)  H3 (n = 6)  H4 (n = 28)  Swing angle,2°  174.6 ± 2.81b (157.7–192.6)  180.1 ± 1.39b (156.4–199.2)  188.3 ± 1.76a (167.8–206.3)  147.6 ± 1.39b (134.1–162.8)  156.0 ± 1.14a (136.5–171.3)  153.8 ± 3.28ab (148.8–158.8)  153.1 ± 1.51a (136.1–171.4)  Foot height, m  0.025 ± 0.0031c (0.013–0.043)  0.040 ± 0.0015b (0.005–0.065)  0.047 ± 0.0019a (0.027–0.073)  0.048 ± 0.0024a (0.018–0.091)  0.032 ± 0.0020b (0.009–0.055)  0.024 ± 0.0057b (0.005–0.059)  0.030 ± 0.0026b (0.005–0.050)  Stride length, m  0.758 ± 0.0121b (0.563–0.873)  0.865 ± 0.0112a (0.669–1.055)  0.809 ± 0.0138b (0.652–0.901)  0.850 ± 0.0179a (0.667–1.078)  0.829 ± 0.0161a (0.565–1.062)  0.684 ± 0.0458b (0.618–0.814)  0.815 ± 0.0192a (0.559–1.034)  Stance time, ms  873.0 ± 33.63a (695.5–1277.2)  633.8 ± 16.93b (344.7–854.1)  571.1 ± 21.30c (335.0–839.9)  602.5 ± 18.65b (395.0–727.6)  571.0 ± 17.03b (282.1–737.3)  637.9 ± 48.15ab (583.8–791.3)  755.0 ± 20.04a (595.3–1175.1)  Percentage of time WS,3 %  71.3 ± 1.58ab (61.9–85.4)  72.2 ± 1.20a (56.7–86.2)  68.4 ± 1.30b (44.7–76.4)  50.7 ± 2.01b (21.8–87.6)  47.7 ± 1.94b (20.2–62.9)  65.8 ± 4.90a (58.2–79.3)  51.5 ± 2.11b (36.9–84.8)  Stance angle,°  207.6 ± 2.07 (193.1–218.3)  211.8 ± 0.99 (195.2–224.6)  208.7 ± 1.27 (193.2–222.1)  –  –  –  –  Step, per min  –  –  –  5.3 ± 0.56b (0.8–17.7)  4.8 ± 0.51b (1.0–16.3)  24.3 ± 1.44a (15.8–35.4)  4.3 ± 0.60b (1.5–9.0)    Fore legs  Hind legs  Variable  F1 (n = 14)1  F2 (n = 65)  F3 (n = 39)  H1 (n = 33)  H2 (n = 50)  H3 (n = 6)  H4 (n = 28)  Swing angle,2°  174.6 ± 2.81b (157.7–192.6)  180.1 ± 1.39b (156.4–199.2)  188.3 ± 1.76a (167.8–206.3)  147.6 ± 1.39b (134.1–162.8)  156.0 ± 1.14a (136.5–171.3)  153.8 ± 3.28ab (148.8–158.8)  153.1 ± 1.51a (136.1–171.4)  Foot height, m  0.025 ± 0.0031c (0.013–0.043)  0.040 ± 0.0015b (0.005–0.065)  0.047 ± 0.0019a (0.027–0.073)  0.048 ± 0.0024a (0.018–0.091)  0.032 ± 0.0020b (0.009–0.055)  0.024 ± 0.0057b (0.005–0.059)  0.030 ± 0.0026b (0.005–0.050)  Stride length, m  0.758 ± 0.0121b (0.563–0.873)  0.865 ± 0.0112a (0.669–1.055)  0.809 ± 0.0138b (0.652–0.901)  0.850 ± 0.0179a (0.667–1.078)  0.829 ± 0.0161a (0.565–1.062)  0.684 ± 0.0458b (0.618–0.814)  0.815 ± 0.0192a (0.559–1.034)  Stance time, ms  873.0 ± 33.63a (695.5–1277.2)  633.8 ± 16.93b (344.7–854.1)  571.1 ± 21.30c (335.0–839.9)  602.5 ± 18.65b (395.0–727.6)  571.0 ± 17.03b (282.1–737.3)  637.9 ± 48.15ab (583.8–791.3)  755.0 ± 20.04a (595.3–1175.1)  Percentage of time WS,3 %  71.3 ± 1.58ab (61.9–85.4)  72.2 ± 1.20a (56.7–86.2)  68.4 ± 1.30b (44.7–76.4)  50.7 ± 2.01b (21.8–87.6)  47.7 ± 1.94b (20.2–62.9)  65.8 ± 4.90a (58.2–79.3)  51.5 ± 2.11b (36.9–84.8)  Stance angle,°  207.6 ± 2.07 (193.1–218.3)  211.8 ± 0.99 (195.2–224.6)  208.7 ± 1.27 (193.2–222.1)  –  –  –  –  Step, per min  –  –  –  5.3 ± 0.56b (0.8–17.7)  4.8 ± 0.51b (1.0–16.3)  24.3 ± 1.44a (15.8–35.4)  4.3 ± 0.60b (1.5–9.0)  a–cWithin a row and a type of legs, means without a common superscript differ (P < 0.05). 1Number of legs. 2Carpal angle for fore legs and tarsal angle for hind legs. 3WS = weight shifting. View Large Table 6. Categorization of fore legs into 3 groups (F1, F2 and F3) and hind legs into 4 groups (H1, H2, H3 and H4) according to cluster analysis using 6 variables from kinematics, force plate, and accelerometer methods. Values are presented as least square means ± SEM (minimum–maximum). The leg is the experimental unit.   Fore legs  Hind legs  Variable  F1 (n = 14)1  F2 (n = 65)  F3 (n = 39)  H1 (n = 33)  H2 (n = 50)  H3 (n = 6)  H4 (n = 28)  Swing angle,2°  174.6 ± 2.81b (157.7–192.6)  180.1 ± 1.39b (156.4–199.2)  188.3 ± 1.76a (167.8–206.3)  147.6 ± 1.39b (134.1–162.8)  156.0 ± 1.14a (136.5–171.3)  153.8 ± 3.28ab (148.8–158.8)  153.1 ± 1.51a (136.1–171.4)  Foot height, m  0.025 ± 0.0031c (0.013–0.043)  0.040 ± 0.0015b (0.005–0.065)  0.047 ± 0.0019a (0.027–0.073)  0.048 ± 0.0024a (0.018–0.091)  0.032 ± 0.0020b (0.009–0.055)  0.024 ± 0.0057b (0.005–0.059)  0.030 ± 0.0026b (0.005–0.050)  Stride length, m  0.758 ± 0.0121b (0.563–0.873)  0.865 ± 0.0112a (0.669–1.055)  0.809 ± 0.0138b (0.652–0.901)  0.850 ± 0.0179a (0.667–1.078)  0.829 ± 0.0161a (0.565–1.062)  0.684 ± 0.0458b (0.618–0.814)  0.815 ± 0.0192a (0.559–1.034)  Stance time, ms  873.0 ± 33.63a (695.5–1277.2)  633.8 ± 16.93b (344.7–854.1)  571.1 ± 21.30c (335.0–839.9)  602.5 ± 18.65b (395.0–727.6)  571.0 ± 17.03b (282.1–737.3)  637.9 ± 48.15ab (583.8–791.3)  755.0 ± 20.04a (595.3–1175.1)  Percentage of time WS,3 %  71.3 ± 1.58ab (61.9–85.4)  72.2 ± 1.20a (56.7–86.2)  68.4 ± 1.30b (44.7–76.4)  50.7 ± 2.01b (21.8–87.6)  47.7 ± 1.94b (20.2–62.9)  65.8 ± 4.90a (58.2–79.3)  51.5 ± 2.11b (36.9–84.8)  Stance angle,°  207.6 ± 2.07 (193.1–218.3)  211.8 ± 0.99 (195.2–224.6)  208.7 ± 1.27 (193.2–222.1)  –  –  –  –  Step, per min  –  –  –  5.3 ± 0.56b (0.8–17.7)  4.8 ± 0.51b (1.0–16.3)  24.3 ± 1.44a (15.8–35.4)  4.3 ± 0.60b (1.5–9.0)    Fore legs  Hind legs  Variable  F1 (n = 14)1  F2 (n = 65)  F3 (n = 39)  H1 (n = 33)  H2 (n = 50)  H3 (n = 6)  H4 (n = 28)  Swing angle,2°  174.6 ± 2.81b (157.7–192.6)  180.1 ± 1.39b (156.4–199.2)  188.3 ± 1.76a (167.8–206.3)  147.6 ± 1.39b (134.1–162.8)  156.0 ± 1.14a (136.5–171.3)  153.8 ± 3.28ab (148.8–158.8)  153.1 ± 1.51a (136.1–171.4)  Foot height, m  0.025 ± 0.0031c (0.013–0.043)  0.040 ± 0.0015b (0.005–0.065)  0.047 ± 0.0019a (0.027–0.073)  0.048 ± 0.0024a (0.018–0.091)  0.032 ± 0.0020b (0.009–0.055)  0.024 ± 0.0057b (0.005–0.059)  0.030 ± 0.0026b (0.005–0.050)  Stride length, m  0.758 ± 0.0121b (0.563–0.873)  0.865 ± 0.0112a (0.669–1.055)  0.809 ± 0.0138b (0.652–0.901)  0.850 ± 0.0179a (0.667–1.078)  0.829 ± 0.0161a (0.565–1.062)  0.684 ± 0.0458b (0.618–0.814)  0.815 ± 0.0192a (0.559–1.034)  Stance time, ms  873.0 ± 33.63a (695.5–1277.2)  633.8 ± 16.93b (344.7–854.1)  571.1 ± 21.30c (335.0–839.9)  602.5 ± 18.65b (395.0–727.6)  571.0 ± 17.03b (282.1–737.3)  637.9 ± 48.15ab (583.8–791.3)  755.0 ± 20.04a (595.3–1175.1)  Percentage of time WS,3 %  71.3 ± 1.58ab (61.9–85.4)  72.2 ± 1.20a (56.7–86.2)  68.4 ± 1.30b (44.7–76.4)  50.7 ± 2.01b (21.8–87.6)  47.7 ± 1.94b (20.2–62.9)  65.8 ± 4.90a (58.2–79.3)  51.5 ± 2.11b (36.9–84.8)  Stance angle,°  207.6 ± 2.07 (193.1–218.3)  211.8 ± 0.99 (195.2–224.6)  208.7 ± 1.27 (193.2–222.1)  –  –  –  –  Step, per min  –  –  –  5.3 ± 0.56b (0.8–17.7)  4.8 ± 0.51b (1.0–16.3)  24.3 ± 1.44a (15.8–35.4)  4.3 ± 0.60b (1.5–9.0)  a–cWithin a row and a type of legs, means without a common superscript differ (P < 0.05). 1Number of legs. 2Carpal angle for fore legs and tarsal angle for hind legs. 3WS = weight shifting. View Large Fore Legs. Legs in the category F1 were characterized by having a low foot height (P = 0.001) and a high stance time (P = 0.0002). Legs in the category F2 were characterized by having a high stride length (P = 0.0022) and a high percentage of time WS (P = 0.0114). Legs in the category F3 were characterized by having a wide carpal angle during swing phase (P = 0.0043), a high foot height (P = 0.0011), a short stance time (P = 0.011), and a low percentage of time WS (P = 0.0114). Sows having both fore legs in category F2 had a lower contralateral ratio than sows having both fore legs in category F3 (F1F1: 0.70 [0.64–0.76]; F2F2: 0.63 [0.60–0.65]; and F3F3: 0.74 [0.71–0.76]; P < 0.0001). A low contralateral ratio reflects an asymmetry of weight bearing between pairs of legs, which is consistent with the fact that legs in F2 had a high percentage of time WS compared to F3 (Table 6). Six combinations of fore leg types were observed: F1F1 (n = 4 sows), F2F1 (n = 3 sows), F2F2 (n = 27 sows), F2F3 (n = 5 sows), F3F1 (n = 1 sows), and F3F3 (n = 18 sows). Therefore, most of the sows had their fore legs within the same category, with F2 and F3 being the most represented. When looking at the distribution of sow within these combinations according to the presence or absence of a problem (from the detailed visual gait evaluation) on fore legs during the lameness score (Table 7), the relationship was not clear. Among sows having no problem on the fore legs, 56.8% were F2F2 and 27.0% were F3F3. But when a problem was noted on the fore legs, 35.0% of sows were F3F3 and 30.0% were F2F2. Therefore, it is difficult to clearly classify sows as lame or not based on fore leg categories. Table 7. Percentage of sows within categories from the cluster analysis in relation to the presence or absence of a gait problem identified by the visual detailed gait observation. Categories are F1, F2, F3 for fore legs and H1, H2, H3, H4 for hind legs   Combinations for both fore legs  Combinations for both hind legs  Leg problem  n1  F1F1 (n = 4)  F2F1 (n = 3)  F2F2 (n = 27)  F2F3 (n = 5)  F3F1 (n = 1)  F3F3 (n = 18)  n  H1H1 (n = 8)  H1H2 (n = 8)  H2H2 (n = 18)  H3H1 (n = 1)  H3H3 (n = 2)  H4H1 (n = 8)  H4H2 (n = 6)  H4H4 (n = 7)  Absence  37  5.4  0  56.8  8.1  2.7  27.0  18  11.1  27.8  33.3  0  0  16.7  5.6  5.6  Presence  20  10.0  15.0  30.0  10.0  0  35.0  33  18.2  9.1  27.3  3.0  6.1  9.1  12.1  15.2    Combinations for both fore legs  Combinations for both hind legs  Leg problem  n1  F1F1 (n = 4)  F2F1 (n = 3)  F2F2 (n = 27)  F2F3 (n = 5)  F3F1 (n = 1)  F3F3 (n = 18)  n  H1H1 (n = 8)  H1H2 (n = 8)  H2H2 (n = 18)  H3H1 (n = 1)  H3H3 (n = 2)  H4H1 (n = 8)  H4H2 (n = 6)  H4H4 (n = 7)  Absence  37  5.4  0  56.8  8.1  2.7  27.0  18  11.1  27.8  33.3  0  0  16.7  5.6  5.6  Presence  20  10.0  15.0  30.0  10.0  0  35.0  33  18.2  9.1  27.3  3.0  6.1  9.1  12.1  15.2  1Number of sows. View Large Table 7. Percentage of sows within categories from the cluster analysis in relation to the presence or absence of a gait problem identified by the visual detailed gait observation. Categories are F1, F2, F3 for fore legs and H1, H2, H3, H4 for hind legs   Combinations for both fore legs  Combinations for both hind legs  Leg problem  n1  F1F1 (n = 4)  F2F1 (n = 3)  F2F2 (n = 27)  F2F3 (n = 5)  F3F1 (n = 1)  F3F3 (n = 18)  n  H1H1 (n = 8)  H1H2 (n = 8)  H2H2 (n = 18)  H3H1 (n = 1)  H3H3 (n = 2)  H4H1 (n = 8)  H4H2 (n = 6)  H4H4 (n = 7)  Absence  37  5.4  0  56.8  8.1  2.7  27.0  18  11.1  27.8  33.3  0  0  16.7  5.6  5.6  Presence  20  10.0  15.0  30.0  10.0  0  35.0  33  18.2  9.1  27.3  3.0  6.1  9.1  12.1  15.2    Combinations for both fore legs  Combinations for both hind legs  Leg problem  n1  F1F1 (n = 4)  F2F1 (n = 3)  F2F2 (n = 27)  F2F3 (n = 5)  F3F1 (n = 1)  F3F3 (n = 18)  n  H1H1 (n = 8)  H1H2 (n = 8)  H2H2 (n = 18)  H3H1 (n = 1)  H3H3 (n = 2)  H4H1 (n = 8)  H4H2 (n = 6)  H4H4 (n = 7)  Absence  37  5.4  0  56.8  8.1  2.7  27.0  18  11.1  27.8  33.3  0  0  16.7  5.6  5.6  Presence  20  10.0  15.0  30.0  10.0  0  35.0  33  18.2  9.1  27.3  3.0  6.1  9.1  12.1  15.2  1Number of sows. View Large Hind Legs. Legs in the category H1 were characterized by a low swing tarsal angle (P = 0.0012) and a high foot height (P < 0.0001). Legs in category H3 were characterized by a low stride length (P = 0.0035), a high percentage of time WS (P = 0.0081), and a high number of steps per minute (P < 0.0001). Legs in category H4 were characterized by a long stance time (P < 0.0001). Category H2 grouped legs having medium values for each variable. Sows having both legs in category H3 had the lowest contralateral ratio compared to sows having both legs in other categories (H1H1: 0.73 [0.65–0.81]; H2H2: 0.75 [0.70–0.81]; H3H3: 0.37 [0.30–0.46]; and H4H4: 0.63 [0.56–0.72]; P < 0.0001). This is in agreement with the fact that the category H3 had a high percentage of time WS (Table 6). Eight combinations of hind leg types were observed: H1H1 (n = 8 sows), H1H2 (n = 8 sows), H2H2 (n = 18 sows), H3H1 (n = 1 sow), H3H3 (n = 2 sows), H4H1 (n = 8 sows), H4H2 (n = 6 sows), and H4H4 (n = 7 sows). The presence or absence of a problem on hind legs does not appear to be associated with combinations of leg types (Table 7). When looking at the classification of each of the 4 legs per sow within categories, 30 sows had their 2 fore legs in the same categories and their 2 hind legs in the same categories. Eighteen sows had their fore legs in the same categories but hind legs in different categories. Five sows had their fore legs in different categories but hind legs in the same. Only 4 sows had their 4 legs in 4 different categories. It appeared, therefore, that, in general, pairs of legs presented the same characteristics. Results showed that the categorization using automated methods identified 8 types of sows (with at least 3 sows per type) and 17 sows with unique characteristics. To conclude, results from the categorization of sows in this study showed a great variety of types of sows and thus underscored the difficulty finding sows with similar characteristics. Conclusion The dissociation between lameness at walk and at a static posture as well as the multiple causes of lameness can bring complication in determining a relationship between general visual lameness score and variables from automated methods. Therefore, only variables that showed differences according to general lameness score, such as contralateral ratio and frequency of WS from the force plate, tarsal angle and stance time from the kinematics, and stepping from accelerometers, can be considered good indicators of lameness. Based on the results from this study, it would appear that the simplified 3-point visual score corresponds more to different types of lameness rather than to a scale of severity, with score 1 possibly more related to gait alteration (identified by the kinematic method) while score 2 was more related to to weight-bearing problems (identified by the force plate). Further research working with animals with an identified lameness problem beforehand or longitudinal studies need to be undertaken to correlate a measure from automated methods to a type of lameness and to determine its threshold value for an accurate diagnosis. This type of longitudinal research has been undertaken in induced-lame sows (Abell et al., 2014) and needs to be developed on sows with naturally occurring lameness to reflect the variety of lameness problems. LITERATURE CITED Abell C. E. Johnson A. K. Karriker L. A. Rothschild M. F. Hoff S. J. Sun G. Fitzgerald R. F. Stalder K. J. 2014. Using classification trees to detect induced sow lameness with a transient model. Animal  8( 6): 1000– 1009. Google Scholar CrossRef Search ADS PubMed  Agriculture and Agri-Food Canada 1993. Recommended code of practice for the care and handling of farm animals—Pigs. Publ. no. 1898E. Agriculture and Agri-Food Canada, Ottawa, ON, Canada. Bonde M. Rousing T. Badsberg J. H. Sørensen J. T. 2004. Associations between lying-down behaviour problems and body condition, limb disorders and skin lesions of lactating sows housed in farrowing crates in commercial sow herds. Livest. Prod. Sci.  87: 179– 187. Google Scholar CrossRef Search ADS   Canadian Council on Animal Care (CCAC) 2009. Guidelines on the care and use of farm animals in research, teaching and testing. CCAC, Ottawa, ON, Canada. Chapinal N. de Passillé A. M. Rushen J. 2009. Weight distribution and gait in cattle are affected by milking and late pregnancy. J. Dairy Sci.  92: 581– 588. Google Scholar CrossRef Search ADS PubMed  Chapinal N. de Passillé A. M. Rushen J. Wagner S. 2010. Automated methods for detecting lameness and measuring analgesia in dairy cattle. J. Dairy Sci.  93: 2007– 2013. Google Scholar CrossRef Search ADS PubMed  Conte S. Bergeron R. Grégoire J. Gète M. D'Allaire S. Meunier-Salaün M. C. Devillers N. 2014. On-farm evaluation of methods to assess welfare of gestating sows. Animal.  8: 1153– 1161. Google Scholar CrossRef Search ADS PubMed  D'Eath R. B. 2012. Repeated locomotion scoring of a sow herd to measure lameness: Consistency over time, the effect of sow characteristics and inter-observer reliability. Anim. Welf.  21: 219– 231. Google Scholar CrossRef Search ADS   Flower F. C. Sanderson D. J. Weary D. M. 2005. Hoof pathologies influence kinematic measures of dairy cow gait. J. Dairy Sci.  88: 3166– 3173. Google Scholar CrossRef Search ADS PubMed  Grégoire J. Bergeron R. D'Allaire S. Meunier-Salaün M.-C. Devillers N. 2013. Assessment of lameness in sows using gait, footprints, postural behaviour and foot lesion analysis. Animal  7: 1163– 1173. Google Scholar CrossRef Search ADS PubMed  Karriker L. A. Abell C. E. Pairis M. D. Holt W. A. Sun G. Coetzee J. F. Johnson A. K. Hoff S. J. Stalder K. J. 2013. Validation of a lameness model in sows using physiological and mechanical measurements. J. Anim. Sci.  91: 130– 136. Google Scholar CrossRef Search ADS PubMed  KilBride A. L. Gillman C. E. Green L. E. 2010. A cross sectional study of prevalence and risk factors for foot lesions and abnormal posture in lactating sows on commercial farms in England. Anim. Welf.  19: 473– 480. Main D. C. J. Clegg J. Spatz A. Green L. E. 2000. Repeatability of a lameness scoring system for finishing pigs. Vet. Rec.  147: 574– 576. Google Scholar CrossRef Search ADS PubMed  Mohling C. M. Johnson A. K. Coetzee J. F. Karriker L. A. Abell C. E. Millman S. T. Stalder K. J. 2014. Kinematics as objective tools to evaluate lameness phases in multiparous sows. Livest. Sci.  165: 120– 128. Google Scholar CrossRef Search ADS   Pastell M. Hänninen L. de Passillé A. M. Rushen J. 2010. Measures of weight distribution of dairy cows to detect lameness and the presence of hoof lesions. J. Dairy Sci.  93: 954– 960. Google Scholar CrossRef Search ADS PubMed  Pastell M. Takko H. Gröhn H. Hautala M. Poikalainen V. Praks J. Veermäe I. Kujala M. Ahokas J. 2006. Assessing cows' welfare: Weighing the cow in a milking robot. Biosystems Eng.  93: 81– 87. Google Scholar CrossRef Search ADS   Pastell M. Tiusanen J. Hakojärvi M. Hänninen L. 2009. A wireless accelerometer system with wavelet analysis for assessing lameness in cattle. Biosystems Eng.  104: 545– 551. Google Scholar CrossRef Search ADS   Pluym L. Van Nuffel A. Dewulf J. Cools A. Vangroenweghe F. Van Hoorebeke S. Maes D. 2011. Prevalence and risk factors of claw lesions and lameness in pregnant sows in two types of group housing. Vet. Med.  56: 101– 109. Pluym L. M. Maes D. Vangeyte J. Mertens K. Baert J. Van Weyenberg S. Millet S. Van Nuffel A. 2013. Development of a system for automatic measurements of force and visual stance variables for objective lameness detection in sows: SowSIS. Biosystems Eng.  116: 64– 74. Google Scholar CrossRef Search ADS   Ringgenberg N. Bergeron R. Devillers N. 2010. Validation of accelerometers to automatically record sow postures and stepping behaviour. Appl. Anim. Behav. Sci.  128: 37– 44. Google Scholar CrossRef Search ADS   Rushen J. Pombourcq E. de Passillé A. M. 2007. Validation of two measures of lameness in dairy cows. Appl. Anim. Behav. Sci.  106: 173– 177. Google Scholar CrossRef Search ADS   Sun G. Fitzgerald R. F. Stalder K. J. Karriker L. A. Johnson A. K. Hoff S. J. 2011. Development of an embedded microcomputer-based force plate system for measuring sow weight distribution and detection of lameness. Appl. Eng. Agric.  27: 475– 482. Google Scholar CrossRef Search ADS   Thorup V. M. Tøgersen F. A. Jørgensen B. Jensen B. R. 2007. Biomechanical gait analysis of pigs walking on solid concrete floor. Animal  1: 708– 715. Google Scholar CrossRef Search ADS PubMed  Veissier I. Jensen K. K. Botreau R. Sandoe P. 2011. Highlighting ethical decisions underlying the scoring of animal welfare in the Welfare Quality@ scheme. Anim. Welf.  20: 89– 101. von Wachenfelt H. Pinzke S. Nilsson C. Olsson O. Ehlorsson C.-J. 2009. Force analysis of unprovoked pig gait on clean and fouled concrete surfaces. Biosystems Eng.  104: 250– 257. Google Scholar CrossRef Search ADS   Wells G. A. H. 1984. Farm practice: Locomotor disorders of the pig. In Pract.  6: 43– 53. Google Scholar CrossRef Search ADS   American Society of Animal Science
TI - Measure and characterization of lameness in gestating sows using force plate, kinematic, and accelerometer methods
JF - Journal of Animal Science
DO - 10.2527/jas.2014-7865
DA - 2014-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/measure-and-characterization-of-lameness-in-gestating-sows-using-force-nt3CuJtDI8
SP - 5693
EP - 5703
VL - 92
IS - 12
DP - DeepDyve
ER -