TY - JOUR
AU1 - Salak-Johnson, J. L.
AU2 - DeDecker, A. E.
AU3 - Levitin, H. A.
AU4 - McGarry, B. M.
AB - Abstract Limited space allowance within the standard gestation stall is an important welfare concern because it restricts the ability of the sow to make postural adjustments and hinders her ability to perform natural behaviors. Therefore, we evaluated the impacts of increasing stall space and/or providing sows the freedom to access a small pen area on sow well-being using multiple welfare metrics. A total of 96 primi- and multiparous crossbred sows were randomly assigned in groups of 4 sows/treatment across 8 replicates to 1 of 3 stall treatments (TRT): standard stall (CTL; dimensions: 61 by 216 cm), width-adjustable stall (flex stall [FLX]; dimensions: adjustable width of 56 to 79 cm by 216 cm), or an individual walk-in/lock-in stall with access to a small communal open-pen area at the rear of the stall (free-access stall [FAS]; dimensions: 69 by 226 cm). Lesion scores, behavior, and immune and productivity traits were measured at various gestational days throughout the study. Total lesion scores were greatest for sows in FAS and least for sows in FLX (P < 0.001). Higher-parity sows in FAS had the most severe lesion scores (TRT × parity, P < 0.0001) and scores were greatest at all gestational days (TRT × day, P < 0.05). Regardless of parity, sows in FLX had the least severe scores (P < 0.0001). As pregnancy progressed, lesion scores increased among sows in CTL (P < 0.05). Sow BW and backfat (BF) were greater for sows in FLX and FAS (P < 0.05), and BCS and BF were greater for parity 1 and 2 sows in FAS than the same parity sows in CTL (TRT × parity, P < 0.05). Duration and frequency of some postural behaviors and sham chew behavior were affected by TRT (P < 0.05) and time of day (TRT × day, P < 0.05). These data indicate that adequate stall space, especially late in gestation, may improve the well-being of higher-parity and heavier-bodied gestating sows as assessed by changes in postural behaviors, lesion severity scores, and other sow traits. Moreover, compromised welfare measures found among sows in various stall environments may be partly attributed to the specific constraints of each stall system such as restricted stall space in CTL, insufficient floor space in the open-pen area of the FAS system, and gate design of the FLX (e.g., direction of bars and feeder space). These results also indicate that parity and gestational day are additional factors that may exacerbate the effects of restricted stall space or insufficient pen space, further compromising sow well-being. INTRODUCTION The most critical welfare issue facing the swine industry is the use of the individual gestation stall (0.61 by 2.13 m). There are both benefits and drawbacks in terms of welfare associated with the stall. The major drawback is the restrictive space allowance, which hinders freedom of movement and ability to perform natural behaviors. In the United States, acceptable requirements for the use of the stall primarily focused on adequate stall space, which allows the sow to easily lie down in full lateral recumbence without simultaneously touching both sides of the stall. The current stall is long and wide enough to accommodate the majority of sows while standing but not lying (McGlone et al., 2004a). The body size of the sow has increased due to genetic selection (Whittemore, 1994), but other factors (e.g., parity, BW) also influence her size (McGlone et al., 2004a). Restricted stall space may adversely affect sow well-being (Anil et al., 2002a, 2006; McGlone et al., 2004a; Anil et al., 2006), specifically postural behaviors among larger sows (Anil et al., 2002b, 2006; Li and Gonyou, 2007). As pregnancy progresses, time spent lying increases and activity decreases (Anil et al., 2006). Moreover, dominant, heavier-bodied sows (DeDecker, 2011) and parities 3 and 4 sows (Rioja-Lang et al., 2013) kept in a housing system with access to both a stall and an open-pen area spend less time in the stall and more time in the open-pen area. Despite the increase in sow body size and greater body size variation industry-wide (Mousten et al., 2011), the standard gestation stall has not been modified to meet the physical or behavioral needs of the modern sow. Therefore, the primary objective of the study was to evaluate the impact of keeping sows in width-adjustable or walk-in/lock-in stalls with access to a small pen on sow well-being using multiple welfare metrics and, moreover, to assess the impact of parity and gestational day on dry sows kept in these different stall environments as compared with sows in standard stalls. MATERIALS AND METHODS Animals, Housing, and Experimental Design The study was conducted at the University of Illinois Imported Swine Research Laboratory (Champaign, IL), and all protocols were approved by the Institutional Animal Care and Use Committee. A total of 96 (n = 12 sows/block) Yorkshire × Landrace crossbred sows were randomly allotted to 1 of 3 stall housing treatments (TRT): a standard stall (CTL), a width-adjustable stall (flex stall [FLX]), or a walk-in/lock-in stall with access to a small pen area shared by 3 other sows (free-access stall [FAS]). All TRT were balanced for BW and parity (parities 0 to 4) and sow parity was equally distributed across all 3 stall housing TRT within block. Sow parity distribution was parity 0 (n = 12), parity 1 and 2 (n = 62), and parity 3 and 4 (n = 22), respectively, and these parity groups were combined based on earlier studies (Salak-Johnson et al., 2007). The initial location of all 3 housing TRT were randomly allotted in groups of 4 stall spaces per TRT, in the same row, and remained in the same location across all 8 blocks. Sows remained in their respective stall housing TRT until gestational d 110, and then all sows were moved to a common farrowing facility. Pregnancy was diagnosed using a Pharvision B-mode ultrasound machine (AV 2100V; Ambisea Technology Corp., Hong Kong, China) for transabdominal examination. All gilts and sows were kept in conventional stalls until confirmed pregnant at d 30 ± 2 after breeding and then moved to their respective stall TRT space. Stall length for both the CTL and FLX was fixed at 216 cm. The width of the CTL was fixed at 61 cm whereas the width of the FLX could be adjusted from 56 to 79 cm. At d 30 after breeding, the initial width of the FLX was determined by using a cloth measuring tape to measure individual sow dimensions, which were taken from the top of the back to the bottom of the udder (depth) and across the shoulders (width) once the sow was placed in the stall. The initial width of the FLX was adjusted to provide each sow an additional 4 cm (2 cm/side) of space based on individual sow body measurements. The second width adjustment was made at gestational d 89 to achieve an additional 8 cm (4 cm/side) of stall width for all sows. The stall width adjustment was previously determined in a pilot study reported by Zverina (2011). The original goal was to adjust the width of the FLX based on midgirth measurements of individual sows, so that each sow could lie in full lateral recumbence without touching the sides of her stall. Moreover, gestational day for the second width adjustment was also standardized based on findings reported by Zverina (2011) in which it was determined that the majority of sows required an increase in stall width at d 90 of gestation. The FLX stalls used in this study were not commercially available and were designed by John Kane, formerly of the University of Wisconsin. The FAS dimensions were 69 by 226 cm for each stall, and the small pen area outside of the 4 stalls was 2.44 by 1.46 m (LAAKE; CTB International Corp., Herzlake, Germany). The flooring within each stall TRT space consisted of 81 cm of solid concrete in the front and 132 cm of slatted concrete in the back of each individual stall. The pen area in the FAS system consisted of 173 cm of slatted flooring, which was continuous from the end of the back gate on the stall, and then 81 cm of solid flooring beyond the slatted portion. Sows were artificially inseminated within 24 h after estrus onset and again 24 h later. Sows were housed in a well-insulated, mechanically ventilated gestation barn and the ambient temperature was set at 22°C but fluctuated ±2°C across seasons; all seasons were replicated. Lighting was set at 12:12 h light:dark schedule, with lights on at 0600 h and off at 1800 h. All sows were individually fed a diet formulated to meet or exceed established nutrient allowances (NRC, 1998). All sows were manually fed daily at 0600 h by the same animal caretaker in the same stall order. Feed was only added to the first stall once all sows in the FAS had entered their stall space and the back gates were closed. Each sow was fed 2.3 ± 0.45 kg/d corn–soy–based diet having a calculated composition (as fed basis) 12.5% CP and providing a ME density of a calculated 3,300 kcal/kg. Each individual stall space contained 1 nipple waterer and no water was available in the pen area of the FAS system. Both the CTL and FAS spaces were fitted with individual feeders, whereas sows housed in FLX were fed on the floor of their stall. The FLX could not be fitted with feeders due to the design of the front gate. Body Condition and Skin Lesion Scores Sow BCS and skin lesion severity score were recorded at d 0 (wean) and gestational d 30, 45, 60, 75, 89, 103, and 110 using a visual appraisal (sow's rear aspect) method described by Coffey et al. (1999) and Salak-Johnson et al. (2007). At weaning, the entire group of sows was scored before moving into the breeding stalls to obtain a baseline score for sows that would be used in the study. Sow BCS (1 = emaciated to 5 = obese) and skin lesion scores were assessed by trained individuals. Specific body locations used to assess skin lesion scores included the head, ears, neck, chest/breast, shoulders, back, udder, rear, vulva, perineum, legs, and hooves (Salak-Johnson et al., 2007). Scoring definitions were 0 = normal/no lesions; 1 = dehairing, callus, balding; 2 = redness, swelling; 3 = swelling plus callus, abscess; 4 = moderate wound, scabbed over scratch; 5 = marked wound, fresh scratch; and 6 = severe wound, open. Sow skin lesion scores were based on the presence or absence of an apparently new or old lesion in conjunction with severity of the wound, and a sow could receive a combined lesion score for a given location on a particular day (Salak-Johnson et al., 2007). Sows received both an average score for each specific body region on a particular gestational day and a total lesion score across all body regions on a particular gestational day. Sow- and Litter-Related Measures Sow BW and backfat (BF) depth were recorded on d 0 (wean), 30, 89, and 110 of gestation. Sow BF was measured using a longitudinal imaging ultrasound scan anterior to the last rib and over the middle of the LM using an Aloka Model 500V B-mode ultrasound scanner fitted with an Aloka 5011 probe (Corometrics Medical Systems, Wallingford, CT). Litter-related traits included total numbers of piglets born alive, stillborn, laid on, and euthanized due to low birth weight or deformation and the number weaned. Litter and mean piglet birth BW, weaning BW, and mean BW gain from birth to weaning were also recorded. Blood Sample Collection and Analysis All sows were nose snared, and approximately 10 mL of blood was collected via jugular venipuncture using Vacutainers (Tyco Healthcare, Mansfield, MA) containing either sodium heparin or EDTA (the procedure lasted <2 min) at d 30, 31, 90, and 110 of gestation at 0700 h for immune traits. An additional blood sample was collected at d 89 of gestation from sows in the FLX before the second stall width adjustment. Blood samples were immediately placed on ice and then centrifuged at 660 × g for 20 min at 4°C, and plasma was removed and stored at –20°C until analysis. Total white blood cell counts were generated electronically using a Coulter Z1 particle counter (Beckman Coulter, Miami, FL) by adding 10 μL of whole blood to Isoflow (10 mL; Beckman Coulter), and red blood cells were lysed with Zap-o-globin (Beckman Coulter). Whole-blood smears were made, fixed in methanol, stained with Hema-3 staining system (Fisher Scientific, Houston, TX), and viewed under a light microscope to determine leukocyte differential counts. A validated commercial RIA (Coat-A-Count; Diagnostic Products, Inc, Los Angeles, CA) was used for porcine plasma cortisol concentration (Sutherland et al., 2005), with an intra- and interassay CV of 4.5 and 8.5%, respectively. Functional Immune Assays: Innate and Adaptive Functional immune assays were used to assess the immune status of pregnant sows as a measure of health and well-being. For all assays, whole blood was diluted with Roswell Park Memorial Institute (RPMI) medium (Gibco, Carlsbad, CA) layered over Histopaque 1077 (density = 1.077 g/mL; Sigma-Aldrich, St. Louis, MO) and 1119 (density = 1.119 g/mL; Sigma-Aldrich) and centrifuged at 700 × g for 30 min at 25°C. Lymphocytes were removed from the 1077 layer and neutrophils were removed from the 1119 layer. Isolated cells were washed twice in RPMI medium, resuspended in RPMI medium, and counted using a cell particle counter (Beckman Coulter). To assess the innate immune status of sows, neutrophil chemotaxis and phagocytosis and natural killer cell cytotoxicity were measured. Briefly, neutrophils were used at a concentration of 3 × 106 cells/mL. Both recombinant human complement-5a (10–7M; Sigma-Aldrich) and recombinant human IL-8 (100 μg/mL; Sigma-Aldrich) were used as chemoattractants. Neutrophil phagocytosis was measured using a flow cytometry–based assay as previously described by Jolie et al. (1997) with minor modifications as described by Niekamp et al. (2007). Fluorescent beads were preincubated for 30 min with non-heat-inactivated porcine serum, and then beads were added to samples at a 10:1 (beads:neutrophils) ratio and samples were incubated for 45 min. The percentage of engulfment of beads by cells was evaluated using a flow cytometer. Natural killer cell cytotoxicity was measured using a commercially available nonradioactive cytotoxicity detection kit (Roche Diagnostics, Indianapolis, IN) as previously described by Sutherland et al. (2005). Briefly, porcine lymphocytes were used as effector cells and K-562 chronic human myelogenous leukemia cells (American Tissue Type Culture Collection, Manassas, VA) were used as target cells. Lymphocytes were adjusted to 1 × 107 cells/mL and K-562 cells adjusted to a constant 10,000 cells per well, and samples were run in triplicate at effector (lymphocytes) to target cell (K-562) ratios of 12.5:1, 25:1, 50:1, and 100:1, respectively. Plates were read using a microplate reader (Bio-Tek Instruments, Winooski, VT) at wavelength 490 nm and reference wavelength 690 nm. Percent cytotoxicity was calculated as described by Lumpkin and McGlone (1992) and an assay was considered valid if maximum release divided by spontaneous release was ≤20%. To assess the adaptive immune status of sows, a mitogen-induced lymphocyte proliferation assay was performed using a CellTiter 96 nonradioactive cell proliferation assay (Promega, Madison, WI) following the manufacturer's protocol with minor modification as previously described by Sutherland et al. (2005). Briefly, porcine lymphocytes were used at a concentration of 5 × 106 cells/mL and placed in triplicate into a sterile 96-well flat-bottom plate. Concanavalin A (ConA; Sigma-Aldrich) and lipopolysaccharide (LPS; Sigma-Aldrich) were used as mitogens (0, 2, and 20 μg/mL) to stimulate T and B cells, respectively. Plates were incubated for 68 h at 37°C in a 5% CO2 humidified incubator and 15 μL dye solution (Promega) was added to each well, and the plates were incubated for 4 h. Solubilization/stop solution (Promega) was added, and the plates were incubated overnight at 37°C and then read using a microplate reader (Bio-Tek Instruments) at wavelength 550 nm with reference wavelength 690 nm. The results are expressed as a proliferation index: optical density(550/690 nm) stimulated cells/optical density (550/690 nm) nonstimulated cells. Behavioral Measures For the comparison of individual sows across all 3 stall housing TRT, continuous sampling of behavior over a 24-h period was collected on d 29, 30, 66, 87, and 102 of gestation for 8 sows per TRT group (2 blocks) for a total of 120 h per sow/TRT. Both frequencies and durations were registered for drink, eat, lay, stand, sit, oral–nasal–facial (ONF), and sham chew behaviors for each sow within stall spaces (Table 1). Data were divided into six 4-h time periods across 24 h: period 1 (0301 to 0700 h), 2 (0701 to 1100 h), 3 (1101 to 1500 h), 4 (1501 to 1900 h), 5 (1901 to 2300 h), and 6 (2301 to 0300 h). Sow behavior was captured using EverFocus black and white surveillance cameras (EverFocus Corp., Duarte, CA). Cameras were mounted on the ceiling and positioned over the stall spaces so that 1 camera could video record 4 sow spaces per housing TRT. Additional behaviors in the pen area of the FAS system were observed and registered for these sows, but findings are reported elsewhere because the pen per se was not the focus of the study (DeDecker, 2011). Table 1. Definitions of each observed and registered behavior Behavior  Description  Lay  Not supported by any limbs. Reclining in ventral or lateral position, in full contact with floor.  Sit  Supported primarily by rear and hind legs with front legs extended  Stand  Body is supported by all 4 legs in contact with floor.  Eat  When feed is present, snout/mouth is in contact with feed.  Drink  Snout/mouth is in contact with nipple waterer.  Oral–nasal–facial  Snout or mouth is in contact with any object besides food or water.  Sham chew  Mouth empty while moving jaw in a repetitive chewing motion  Behavior  Description  Lay  Not supported by any limbs. Reclining in ventral or lateral position, in full contact with floor.  Sit  Supported primarily by rear and hind legs with front legs extended  Stand  Body is supported by all 4 legs in contact with floor.  Eat  When feed is present, snout/mouth is in contact with feed.  Drink  Snout/mouth is in contact with nipple waterer.  Oral–nasal–facial  Snout or mouth is in contact with any object besides food or water.  Sham chew  Mouth empty while moving jaw in a repetitive chewing motion  View Large Table 1. Definitions of each observed and registered behavior Behavior  Description  Lay  Not supported by any limbs. Reclining in ventral or lateral position, in full contact with floor.  Sit  Supported primarily by rear and hind legs with front legs extended  Stand  Body is supported by all 4 legs in contact with floor.  Eat  When feed is present, snout/mouth is in contact with feed.  Drink  Snout/mouth is in contact with nipple waterer.  Oral–nasal–facial  Snout or mouth is in contact with any object besides food or water.  Sham chew  Mouth empty while moving jaw in a repetitive chewing motion  Behavior  Description  Lay  Not supported by any limbs. Reclining in ventral or lateral position, in full contact with floor.  Sit  Supported primarily by rear and hind legs with front legs extended  Stand  Body is supported by all 4 legs in contact with floor.  Eat  When feed is present, snout/mouth is in contact with feed.  Drink  Snout/mouth is in contact with nipple waterer.  Oral–nasal–facial  Snout or mouth is in contact with any object besides food or water.  Sham chew  Mouth empty while moving jaw in a repetitive chewing motion  View Large Statistical Analyses The distribution of all traits was tested for normality using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Only behavioral traits exhibited a skewed distribution, and therefore, a natural-logarithmic transformation was applied to these variables to facilitate the interpretation of results. Physiological, behavioral, and productivity data were all analyzed with the linear–mixed model procedure of SAS (SAS Inst. Inc., Cary, NC), with the repeated measure option. The model for sow BW, BCS, BF, and physiological measurements included the fixed effects factors of stall space (3 levels: CTL, FLX, and FAS), parity (3 levels), day of measurement (from gestation d 30 to 110), TRT × parity, TRT × day, parity × day, and the covariate of the corresponding blood measurement at d 30 of gestation (baseline measurement; d 0, which is the start of the experiment) or corresponding BW for BCS and BF measurements. Random effects of block (sow blocks levels 1 through 8) and sow nested within block were included in the model to account for potential environmental and management differences across blocks and the repeated measurements within sow. Two of the space TRT assigned were for individual sows; therefore, the experimental unit to test stall space effect was sow. The heterogeneity of variance structure accommodated for potential differences in variation across space levels including those due to different number of sows per enclosure (stall or pen area in FAS system; Salak-Johnson et al., 2007). The model for behavioral traits was similar to the model used for physiological measurements, except the hour of day (time periods of 4 h each and a total of 6 daily time periods) was added and the covariate d-30 blood measurement was removed. The ordinal nature of the skin lesion score variable was modeled using PROC GLIMMIX instead of PROC MIXED, but the overall model was similar to one used for physiological measures except gestational day varied and the covariate blood measurement was removed. In addition to the overall P-value of the model terms, least squares means were generated for all traits and statistically separated with pairwise t tests (PDIFF option). Significance was set at P ≤ 0.05, whereas trends are discussed at P ≤ 0.10. RESULTS Treatment × Parity Effects Sow-Related Measures and Lesions. Probability values and means for the effects of stall housing TRT and parity on sow-related traits are presented in Table 2. Parity 0 sows kept in FLX had greater mean BCS than did sows kept in CTL, but mean BCS was similar between parity 0 sows kept in FAS and CTL (TRT × parity, P < 0.05; Table 2). Parity 1 and 2 sows kept in FAS had greater mean BCS than did the same parity sows kept in CTL (TRT × parity, P < 0.05; Table 2). Sow mean BCS was similar among parity 3 and 4 sows, regardless of stall TRT. Parity 0 and parity 1 and 2 sows had similar mean BW, regardless of stall TRT (Table 2). Parity 3 and 4 sows kept in FAS tended to have greater mean BW than did the same parity sows kept in CTL, whereas BW was similar between those sows kept in FLX or CTL (TRT × parity, P < 0.10; Table 2). Parity 0 sows kept in FLX and FAS had greater mean BF depth when compared with the same parity sows kept in CTL (Table 2). Parity 1 and 2 sows kept in FAS had greater BF depth than did the same parity sows kept in CTL (TRT × parity, P < 0.001; Table 2). Table 2. Interactive effects of stall housing treatment and parity on mean BCS, BW, backfat depth, and total lesion score for gestating sows   Stall treatment2      Item1  CTL  FLX  FAS  SEp3  P-value4  Mean BCS, 1–5      Parity 0  3.06b  3.60a  3.33ab  0.17  <0.05      Parity 1 and 2  3.11b  3.18ab  3.32a  0.07        Parity 3 and 4  3.10  3.00  3.17  0.13    Mean BW, kg      Parity 0  184.1  194.2  193.5  5.3  0.07      Parity 1 and 2  211.7  208.3  213.5  2.5        Parity 3 and 4  220.6  221.4  235.2  3.9    Mean backfat, mm      Parity 0  16.7b  19.0a  20.8a  0.12  <0.001      Parity 1 and 2  16.7b  17.1b  19.1a  0.05        Parity 3 and 4  15.7  15.0  16.1  0.09    Total lesion score      Parity 0  5.2a  4.3b  7.9a  0.56  <0.0001      Parity 1 and 2  5.7a  4.6b  9.4a  0.23        Parity 3 and 4  7.8b  4.2c  11.5a  0.41      Stall treatment2      Item1  CTL  FLX  FAS  SEp3  P-value4  Mean BCS, 1–5      Parity 0  3.06b  3.60a  3.33ab  0.17  <0.05      Parity 1 and 2  3.11b  3.18ab  3.32a  0.07        Parity 3 and 4  3.10  3.00  3.17  0.13    Mean BW, kg      Parity 0  184.1  194.2  193.5  5.3  0.07      Parity 1 and 2  211.7  208.3  213.5  2.5        Parity 3 and 4  220.6  221.4  235.2  3.9    Mean backfat, mm      Parity 0  16.7b  19.0a  20.8a  0.12  <0.001      Parity 1 and 2  16.7b  17.1b  19.1a  0.05        Parity 3 and 4  15.7  15.0  16.1  0.09    Total lesion score      Parity 0  5.2a  4.3b  7.9a  0.56  <0.0001      Parity 1 and 2  5.7a  4.6b  9.4a  0.23        Parity 3 and 4  7.8b  4.2c  11.5a  0.41    a,bWithin a row, least squares means lacking common superscripted letters differ (P ≤ 0.05). 1Parity 0 = gilts; Parity 1 and 2 = combined first and second parity sows; Parity 3 and 4 = combined third and fourth parity sows based on statistical analysis and model. Least squares means value ± pooled SE. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall. 3SEp = pooled standard error. 4Probability value for the treatment × parity interaction. View Large Table 2. Interactive effects of stall housing treatment and parity on mean BCS, BW, backfat depth, and total lesion score for gestating sows   Stall treatment2      Item1  CTL  FLX  FAS  SEp3  P-value4  Mean BCS, 1–5      Parity 0  3.06b  3.60a  3.33ab  0.17  <0.05      Parity 1 and 2  3.11b  3.18ab  3.32a  0.07        Parity 3 and 4  3.10  3.00  3.17  0.13    Mean BW, kg      Parity 0  184.1  194.2  193.5  5.3  0.07      Parity 1 and 2  211.7  208.3  213.5  2.5        Parity 3 and 4  220.6  221.4  235.2  3.9    Mean backfat, mm      Parity 0  16.7b  19.0a  20.8a  0.12  <0.001      Parity 1 and 2  16.7b  17.1b  19.1a  0.05        Parity 3 and 4  15.7  15.0  16.1  0.09    Total lesion score      Parity 0  5.2a  4.3b  7.9a  0.56  <0.0001      Parity 1 and 2  5.7a  4.6b  9.4a  0.23        Parity 3 and 4  7.8b  4.2c  11.5a  0.41      Stall treatment2      Item1  CTL  FLX  FAS  SEp3  P-value4  Mean BCS, 1–5      Parity 0  3.06b  3.60a  3.33ab  0.17  <0.05      Parity 1 and 2  3.11b  3.18ab  3.32a  0.07        Parity 3 and 4  3.10  3.00  3.17  0.13    Mean BW, kg      Parity 0  184.1  194.2  193.5  5.3  0.07      Parity 1 and 2  211.7  208.3  213.5  2.5        Parity 3 and 4  220.6  221.4  235.2  3.9    Mean backfat, mm      Parity 0  16.7b  19.0a  20.8a  0.12  <0.001      Parity 1 and 2  16.7b  17.1b  19.1a  0.05        Parity 3 and 4  15.7  15.0  16.1  0.09    Total lesion score      Parity 0  5.2a  4.3b  7.9a  0.56  <0.0001      Parity 1 and 2  5.7a  4.6b  9.4a  0.23        Parity 3 and 4  7.8b  4.2c  11.5a  0.41    a,bWithin a row, least squares means lacking common superscripted letters differ (P ≤ 0.05). 1Parity 0 = gilts; Parity 1 and 2 = combined first and second parity sows; Parity 3 and 4 = combined third and fourth parity sows based on statistical analysis and model. Least squares means value ± pooled SE. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall. 3SEp = pooled standard error. 4Probability value for the treatment × parity interaction. View Large Total skin lesion scores were greater for parity 0 and parity 1 and 2 sows kept in FAS when compared with the same parity sows kept in CTL (TRT × parity, P < 0.0001; Table 2). Parity 0 and parity 1 and 2 sows kept in FLX and CTL had similar lesion scores. Total skin lesion scores were greater (P < 0.0001) for parity 3 and 4 sows housed in CTL when compared with the same parity sows housed in FLX. Parity 3 and 4 sows kept in FAS had greater skin lesion scores than did sows kept in CTL (TRT × parity, P < 0.0001; Table 2). At all parities, sows kept in FAS had greater mean skin lesion scores at the shoulders (Fig. 1A) and rear (Fig. 1B) when compared with sows kept in CTL and FLX (TRT × parity, P < 0.0001). Shoulder lesion scores were similar across all parities for sows kept in either CTL or FLX (Fig. 1A). Parity 1 and 2 sows kept in CTL had greater scores at the rear than did the same parity sows kept in FLX (TRT × parity, P < 0.0001; Fig. 1B). Regardless of parity, sows housed in the FLX had least severe total lesion scores (Table 2) and least severe scores at the shoulders (Fig. 1A) and rear (Fig. 1B) when compared with the same parity sows kept in either the CTL or FAS (TRT × parity, P < 0.0001). Figure 1. View largeDownload slide Interactive effects of stall treatments (CTL = standard stall; FLX = flex stall; FAS = free-access stall) and parity on mean (A) shoulder lesion scores (P < 0.0001) and (B) rear lesion scores (P < 0.0001) for gestating sows kept in different housing environments. Both shoulder (A) and rear skin lesion scores (B) were highest for sows kept in FAS compared with sows in CTL. Parity 3 and 4 sows in FAS had the highest scores at shoulders and rear but most severe scores at the rear. Lesions at shoulders and rear were similar between sows in FLX and CTL with the exception that parity 1 and 2 sows kept in CTL had greater scores than those kept in FLX. An asterisk (*) indicates that stall treatments differed (P ≤ 0.05) within sow parity. Figure 1. View largeDownload slide Interactive effects of stall treatments (CTL = standard stall; FLX = flex stall; FAS = free-access stall) and parity on mean (A) shoulder lesion scores (P < 0.0001) and (B) rear lesion scores (P < 0.0001) for gestating sows kept in different housing environments. Both shoulder (A) and rear skin lesion scores (B) were highest for sows kept in FAS compared with sows in CTL. Parity 3 and 4 sows in FAS had the highest scores at shoulders and rear but most severe scores at the rear. Lesions at shoulders and rear were similar between sows in FLX and CTL with the exception that parity 1 and 2 sows kept in CTL had greater scores than those kept in FLX. An asterisk (*) indicates that stall treatments differed (P ≤ 0.05) within sow parity. Litter-Related Measures. Parity 0 sows kept in FAS (14.2 ± 1.6) had larger litters born than did sows kept in CTL (9.2 ± 0.69), whereas parity 3 and 4 sows kept in FLX (10.8 ± 0.88) had larger litters than did sows kept in CTL (7.2 ± 0.97; TRT × parity, P < 0.0001). More piglets tended to be euthanized among parity 0 sows kept in FAS compared with the same parity sows kept in CTL (0.80 ± 0.40; TRT × parity, P < 0.10). Within stall TRT, parity 1 and 2 sows had larger litters born (11.0 ± 0.71) than did parity 3 and 4 sows kept in CTL and parity 0 sows had larger litters than did parity 3 and 4 sows (9.6 ± 1.4) kept in FAS (TRT × parity, P < 0.001). Litter size was similar among sows kept in FLX, regardless of parity (TRT × parity, P < 0.001). Immune Measures. Probability values and means for the effects of stall housing TRT and sow parity on selected immune traits are presented in Table 3. Parity 0 and parity 1 and 2 sows kept in FAS had greater total numbers of lymphocytes than did sows kept in CTL (TRT × parity, P < 0.05). Sows kept in FLX had similar numbers of lymphocytes and neutrophils compared with the same parity sows kept in CTL (TRT × parity, P < 0.05; Table 3). Parity 1 and 2 sows kept in CTL had a greater (P < 0.005) percentage of eosinophils than did sows housed in FAS, whereas sows kept in FLX had a similar percentage of eosinophils compared with sows kept in CTL (Table 3). Parity 3 and 4 sows kept in CTL had greater ConA-induced lymphocyte proliferation when compared with sows kept in either FLX or FAS (TRT × parity, P < 0.0005; Table 3). Table 3. Interactive effects of stall housing treatment and parity on selected immune traits for gestating sows1   Treatment2      Immune item1  CTL  FLX  FAS  SEp3  P-value4  Lymphocyte, 107/mL      Parity 0  2.5b  2.2b  3.2a  0.29  0.03      Parity 1 and 2  2.6b  2.7ab  3.0a  0.11        Parity 3 and 4  2.7  2.4  2.3  0.29    Neutrophil, 106/mL      Parity 0  5.6ab  4.4b  6.3a  0.55  0.05      Parity 1 and 2  4.8ab  4.7b  5.3a  0.22        Parity 3 and 4  5.3  5.0  4.6  0.40    Eosinophils, %      Parity 0  5.5  5.6  4.5  0.96  0.003      Parity 1 and 2  5.6a  6.4a  4.2b  0.39        Parity 3 and 4  4.6  5.4  5.3  0.69    ConA-induced proliferation      Parity 0  1.66  1.37  1.40  0.37  0.0002      Parity 1 and 2  1.49  1.21  1.42  0.17        Parity 3 and 4  3.13a  1.27b  1.21b  0.32      Treatment2      Immune item1  CTL  FLX  FAS  SEp3  P-value4  Lymphocyte, 107/mL      Parity 0  2.5b  2.2b  3.2a  0.29  0.03      Parity 1 and 2  2.6b  2.7ab  3.0a  0.11        Parity 3 and 4  2.7  2.4  2.3  0.29    Neutrophil, 106/mL      Parity 0  5.6ab  4.4b  6.3a  0.55  0.05      Parity 1 and 2  4.8ab  4.7b  5.3a  0.22        Parity 3 and 4  5.3  5.0  4.6  0.40    Eosinophils, %      Parity 0  5.5  5.6  4.5  0.96  0.003      Parity 1 and 2  5.6a  6.4a  4.2b  0.39        Parity 3 and 4  4.6  5.4  5.3  0.69    ConA-induced proliferation      Parity 0  1.66  1.37  1.40  0.37  0.0002      Parity 1 and 2  1.49  1.21  1.42  0.17        Parity 3 and 4  3.13a  1.27b  1.21b  0.32    a,bWithin a row, least squares means lacking common superscripted letters differ (P ≤ 0.05). 1Parity 0 = gilts; Parity 1 and 2 = combined first and second parity sows; Parity 3 and 4 = combined third and fourth parity sows based on statistical analysis and model; ConA = concanavalin A. Least squares means value ± pooled SE. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall. 3SEp = pooled standard error. 4Probability value for the treatment × parity interaction. View Large Table 3. Interactive effects of stall housing treatment and parity on selected immune traits for gestating sows1   Treatment2      Immune item1  CTL  FLX  FAS  SEp3  P-value4  Lymphocyte, 107/mL      Parity 0  2.5b  2.2b  3.2a  0.29  0.03      Parity 1 and 2  2.6b  2.7ab  3.0a  0.11        Parity 3 and 4  2.7  2.4  2.3  0.29    Neutrophil, 106/mL      Parity 0  5.6ab  4.4b  6.3a  0.55  0.05      Parity 1 and 2  4.8ab  4.7b  5.3a  0.22        Parity 3 and 4  5.3  5.0  4.6  0.40    Eosinophils, %      Parity 0  5.5  5.6  4.5  0.96  0.003      Parity 1 and 2  5.6a  6.4a  4.2b  0.39        Parity 3 and 4  4.6  5.4  5.3  0.69    ConA-induced proliferation      Parity 0  1.66  1.37  1.40  0.37  0.0002      Parity 1 and 2  1.49  1.21  1.42  0.17        Parity 3 and 4  3.13a  1.27b  1.21b  0.32      Treatment2      Immune item1  CTL  FLX  FAS  SEp3  P-value4  Lymphocyte, 107/mL      Parity 0  2.5b  2.2b  3.2a  0.29  0.03      Parity 1 and 2  2.6b  2.7ab  3.0a  0.11        Parity 3 and 4  2.7  2.4  2.3  0.29    Neutrophil, 106/mL      Parity 0  5.6ab  4.4b  6.3a  0.55  0.05      Parity 1 and 2  4.8ab  4.7b  5.3a  0.22        Parity 3 and 4  5.3  5.0  4.6  0.40    Eosinophils, %      Parity 0  5.5  5.6  4.5  0.96  0.003      Parity 1 and 2  5.6a  6.4a  4.2b  0.39        Parity 3 and 4  4.6  5.4  5.3  0.69    ConA-induced proliferation      Parity 0  1.66  1.37  1.40  0.37  0.0002      Parity 1 and 2  1.49  1.21  1.42  0.17        Parity 3 and 4  3.13a  1.27b  1.21b  0.32    a,bWithin a row, least squares means lacking common superscripted letters differ (P ≤ 0.05). 1Parity 0 = gilts; Parity 1 and 2 = combined first and second parity sows; Parity 3 and 4 = combined third and fourth parity sows based on statistical analysis and model; ConA = concanavalin A. Least squares means value ± pooled SE. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall. 3SEp = pooled standard error. 4Probability value for the treatment × parity interaction. View Large Treatment × Gestational Day Effects on Lesions A significant TRT × gestational day interaction occurred for total skin lesion scores (Fig. 2). Total skin lesion scores were greater for sows kept in FAS on gestational d 45, 60, 75, and 89 when compared with sows kept in CTL on those days (TRT × gestational d, P < 0.05; Fig. 2). On gestational d 45 and 60, total lesion scores were greater for sows kept in CTL when compared with sows kept in FLX (TRT × gestational d, P < 0.05; Fig. 2). On gestational d 89 and 103, sows kept in CTL and FAS had more severe skin lesion scores than did sows kept in FLX (TRT × gestational d, P < 0.05; Fig. 2). At all gestational days, sows kept in FLX had less severe lesion scores than did sows kept in FAS (Fig. 2). Figure 2. View largeDownload slide Interactive effects of stall treatments (CTL = standard stall; FLX = flex stall; FAS = free-access stall) and gestational day on total mean skin lesion scores (P < 0.05) for gestating sows kept in different housing environments. On gestational d 45, 60, 75, and 89, sows in FAS had the highest total lesion scores when compared with sows in either CTL or FLX. On gestational d 45 and 60 and then again at d 89 and 103, sows in CTL had greater lesion scores than did sows in FLX. An asterisk (*) indicates that stall treatments differed (P ≤ 0.05) for that day of gestation. Figure 2. View largeDownload slide Interactive effects of stall treatments (CTL = standard stall; FLX = flex stall; FAS = free-access stall) and gestational day on total mean skin lesion scores (P < 0.05) for gestating sows kept in different housing environments. On gestational d 45, 60, 75, and 89, sows in FAS had the highest total lesion scores when compared with sows in either CTL or FLX. On gestational d 45 and 60 and then again at d 89 and 103, sows in CTL had greater lesion scores than did sows in FLX. An asterisk (*) indicates that stall treatments differed (P ≤ 0.05) for that day of gestation. Treatment × Time Period Effects on Behavior A TRT × time period interaction occurred for both durations and frequencies of several sow behaviors (Table 4). During time period 3, sows kept in FLX spent more time sitting than did sows kept in CTL, but sows kept in CTL spent more time sitting than did sows kept in FAS (TRT × time period, P < 0.05; Table 4). Sows kept in FAS spent more time lying during time period 5 than did sows kept in CTL, but time spent lying was similar between sows kept in FLX and CTL (TRT × time period, P = 0.05; Table 4). During time period 6, sows kept in FLX spent more time lying than did sows kept in CTL, but sows in CTL spent more time lying than did sows in FAS (TRT × time period, P = 0.05). Only during time periods 1 and 2 did sows kept in FLX stand more often than sows kept in CTL (TRT × time period, P < 0.05; Table 4). Table 4. Interactive effects of stall housing treatment and time period on duration and frequency of behavior for gestating sows1   Time period3      Behavior2  1  2  3  4  5  6  SEp4  P-value5  Duration, min      Lay          CTL  68.2  16.3  35.9  81.9  139.6b  149.5a  9.7  0.05          FLX  48.5  16.6  25.9  90.9  141.4b  158.3a  10.1          FAS  60.2  29.3  42.2  80.9  169.6a  112.7b  9.5      Stand          CTL  26.1  70.9  41.6  15.8  3.71  3.63  6.8  0.92          FLX  17.4  60.7  32.8  7.91  11.2  6.68  7.2          FAS  19.4  64.7  24.3  10.8  6.52  3.16  7.1      Sit          CTL  1.66  1.59  3.52b  1.78  0.37  0.21  0.68  <0.05          FLX  1.22  1.79  6.46a  0.91  0.40  1.18  0.73          FAS  2.24  1.70  1.89b  1.10  0.06  0.30  0.67      ONF          CTL  8.72  13.2  13.4  7.47  2.83  1.84  2.83  0.42          FLX  4.26  9.69  9.00  2.96  10.2  2.11  3.17          FAS  6.25  9.17  5.93  5.41  9.44  4.10  2.84      Sham chew          CTL  3.32  2.53  4.30  2.05  0.63  0.66  0.63  0.30          FLX  2.79  3.42  4.70  2.86  1.95  2.70  0.69          FAS  4.27  3.47  3.09  1.73  0.18  1.42  0.61  Frequency      Lay          CTL  2.93  2.07  3.77  2.37  1.33  0.57  0.43  0.12          FLX  4.60  3.72  4.48  2.08  1.32  0.96  0.47          FAS  2.85  2.94  3.69  2.47  0.66  1.53  0.43      Stand          CTL  2.41b  2.48b  3.27a  1.94  0.87  0.59  0.36  0.01          FLX  4.71a  4.07a  3.63a  1.47  1.07  0.83  0.38          FAS  2.97b  2.84b  2.51b  1.90  0.53  1.19  0.37      Sit          CTL  2.52  1.70  2.01  1.28  0.61  0.13  0.34  0.69          FLX  2.64  2.40  2.56  0.76  0.54  0.48  0.37          FAS  2.44  2.09  2.47  1.01  0.06  0.84  0.35      ONF          CTL  6.69  10.3  9.95  3.35  1.45  0.94  1.33  0.10          FLX  11.5  10.3  9.94  2.34  0.93  0.90  1.45          FAS  9.78  14.4  6.24  3.99  0.72  1.28  1.35      Sham chew          CTL  5.71b  3.78b  6.00b  2.26  0.76  0.36  1.20  0.01          FLX  11.0a  8.37a  12.5a  2.25  0.84  1.29  1.31          FAS  5.34b  6.97a  4.05b  2.65  0.18  1.03  1.19    Time period3      Behavior2  1  2  3  4  5  6  SEp4  P-value5  Duration, min      Lay          CTL  68.2  16.3  35.9  81.9  139.6b  149.5a  9.7  0.05          FLX  48.5  16.6  25.9  90.9  141.4b  158.3a  10.1          FAS  60.2  29.3  42.2  80.9  169.6a  112.7b  9.5      Stand          CTL  26.1  70.9  41.6  15.8  3.71  3.63  6.8  0.92          FLX  17.4  60.7  32.8  7.91  11.2  6.68  7.2          FAS  19.4  64.7  24.3  10.8  6.52  3.16  7.1      Sit          CTL  1.66  1.59  3.52b  1.78  0.37  0.21  0.68  <0.05          FLX  1.22  1.79  6.46a  0.91  0.40  1.18  0.73          FAS  2.24  1.70  1.89b  1.10  0.06  0.30  0.67      ONF          CTL  8.72  13.2  13.4  7.47  2.83  1.84  2.83  0.42          FLX  4.26  9.69  9.00  2.96  10.2  2.11  3.17          FAS  6.25  9.17  5.93  5.41  9.44  4.10  2.84      Sham chew          CTL  3.32  2.53  4.30  2.05  0.63  0.66  0.63  0.30          FLX  2.79  3.42  4.70  2.86  1.95  2.70  0.69          FAS  4.27  3.47  3.09  1.73  0.18  1.42  0.61  Frequency      Lay          CTL  2.93  2.07  3.77  2.37  1.33  0.57  0.43  0.12          FLX  4.60  3.72  4.48  2.08  1.32  0.96  0.47          FAS  2.85  2.94  3.69  2.47  0.66  1.53  0.43      Stand          CTL  2.41b  2.48b  3.27a  1.94  0.87  0.59  0.36  0.01          FLX  4.71a  4.07a  3.63a  1.47  1.07  0.83  0.38          FAS  2.97b  2.84b  2.51b  1.90  0.53  1.19  0.37      Sit          CTL  2.52  1.70  2.01  1.28  0.61  0.13  0.34  0.69          FLX  2.64  2.40  2.56  0.76  0.54  0.48  0.37          FAS  2.44  2.09  2.47  1.01  0.06  0.84  0.35      ONF          CTL  6.69  10.3  9.95  3.35  1.45  0.94  1.33  0.10          FLX  11.5  10.3  9.94  2.34  0.93  0.90  1.45          FAS  9.78  14.4  6.24  3.99  0.72  1.28  1.35      Sham chew          CTL  5.71b  3.78b  6.00b  2.26  0.76  0.36  1.20  0.01          FLX  11.0a  8.37a  12.5a  2.25  0.84  1.29  1.31          FAS  5.34b  6.97a  4.05b  2.65  0.18  1.03  1.19  a,bWithin a column, least squares means lacking common superscripted letters differ (P < 0.05). 1Data represent 96 pregnant sows across 8 replicates, with sow as the experimental unit. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall; ONF = oral–nasal–facial. 3Time periods were divided into six 4-h time periods across 24 h: period 1 (0300 to 0700 h), 2 (0700 to 1100 h), 3 (1100 to 1500 h), 4 (1500 to 1900 h), 5 (1900 to 2300 h), and 6 (2300 to 0300 h). 4SEp = pooled standard error. 5Probability value for the treatment × time period interaction. View Large Table 4. Interactive effects of stall housing treatment and time period on duration and frequency of behavior for gestating sows1   Time period3      Behavior2  1  2  3  4  5  6  SEp4  P-value5  Duration, min      Lay          CTL  68.2  16.3  35.9  81.9  139.6b  149.5a  9.7  0.05          FLX  48.5  16.6  25.9  90.9  141.4b  158.3a  10.1          FAS  60.2  29.3  42.2  80.9  169.6a  112.7b  9.5      Stand          CTL  26.1  70.9  41.6  15.8  3.71  3.63  6.8  0.92          FLX  17.4  60.7  32.8  7.91  11.2  6.68  7.2          FAS  19.4  64.7  24.3  10.8  6.52  3.16  7.1      Sit          CTL  1.66  1.59  3.52b  1.78  0.37  0.21  0.68  <0.05          FLX  1.22  1.79  6.46a  0.91  0.40  1.18  0.73          FAS  2.24  1.70  1.89b  1.10  0.06  0.30  0.67      ONF          CTL  8.72  13.2  13.4  7.47  2.83  1.84  2.83  0.42          FLX  4.26  9.69  9.00  2.96  10.2  2.11  3.17          FAS  6.25  9.17  5.93  5.41  9.44  4.10  2.84      Sham chew          CTL  3.32  2.53  4.30  2.05  0.63  0.66  0.63  0.30          FLX  2.79  3.42  4.70  2.86  1.95  2.70  0.69          FAS  4.27  3.47  3.09  1.73  0.18  1.42  0.61  Frequency      Lay          CTL  2.93  2.07  3.77  2.37  1.33  0.57  0.43  0.12          FLX  4.60  3.72  4.48  2.08  1.32  0.96  0.47          FAS  2.85  2.94  3.69  2.47  0.66  1.53  0.43      Stand          CTL  2.41b  2.48b  3.27a  1.94  0.87  0.59  0.36  0.01          FLX  4.71a  4.07a  3.63a  1.47  1.07  0.83  0.38          FAS  2.97b  2.84b  2.51b  1.90  0.53  1.19  0.37      Sit          CTL  2.52  1.70  2.01  1.28  0.61  0.13  0.34  0.69          FLX  2.64  2.40  2.56  0.76  0.54  0.48  0.37          FAS  2.44  2.09  2.47  1.01  0.06  0.84  0.35      ONF          CTL  6.69  10.3  9.95  3.35  1.45  0.94  1.33  0.10          FLX  11.5  10.3  9.94  2.34  0.93  0.90  1.45          FAS  9.78  14.4  6.24  3.99  0.72  1.28  1.35      Sham chew          CTL  5.71b  3.78b  6.00b  2.26  0.76  0.36  1.20  0.01          FLX  11.0a  8.37a  12.5a  2.25  0.84  1.29  1.31          FAS  5.34b  6.97a  4.05b  2.65  0.18  1.03  1.19    Time period3      Behavior2  1  2  3  4  5  6  SEp4  P-value5  Duration, min      Lay          CTL  68.2  16.3  35.9  81.9  139.6b  149.5a  9.7  0.05          FLX  48.5  16.6  25.9  90.9  141.4b  158.3a  10.1          FAS  60.2  29.3  42.2  80.9  169.6a  112.7b  9.5      Stand          CTL  26.1  70.9  41.6  15.8  3.71  3.63  6.8  0.92          FLX  17.4  60.7  32.8  7.91  11.2  6.68  7.2          FAS  19.4  64.7  24.3  10.8  6.52  3.16  7.1      Sit          CTL  1.66  1.59  3.52b  1.78  0.37  0.21  0.68  <0.05          FLX  1.22  1.79  6.46a  0.91  0.40  1.18  0.73          FAS  2.24  1.70  1.89b  1.10  0.06  0.30  0.67      ONF          CTL  8.72  13.2  13.4  7.47  2.83  1.84  2.83  0.42          FLX  4.26  9.69  9.00  2.96  10.2  2.11  3.17          FAS  6.25  9.17  5.93  5.41  9.44  4.10  2.84      Sham chew          CTL  3.32  2.53  4.30  2.05  0.63  0.66  0.63  0.30          FLX  2.79  3.42  4.70  2.86  1.95  2.70  0.69          FAS  4.27  3.47  3.09  1.73  0.18  1.42  0.61  Frequency      Lay          CTL  2.93  2.07  3.77  2.37  1.33  0.57  0.43  0.12          FLX  4.60  3.72  4.48  2.08  1.32  0.96  0.47          FAS  2.85  2.94  3.69  2.47  0.66  1.53  0.43      Stand          CTL  2.41b  2.48b  3.27a  1.94  0.87  0.59  0.36  0.01          FLX  4.71a  4.07a  3.63a  1.47  1.07  0.83  0.38          FAS  2.97b  2.84b  2.51b  1.90  0.53  1.19  0.37      Sit          CTL  2.52  1.70  2.01  1.28  0.61  0.13  0.34  0.69          FLX  2.64  2.40  2.56  0.76  0.54  0.48  0.37          FAS  2.44  2.09  2.47  1.01  0.06  0.84  0.35      ONF          CTL  6.69  10.3  9.95  3.35  1.45  0.94  1.33  0.10          FLX  11.5  10.3  9.94  2.34  0.93  0.90  1.45          FAS  9.78  14.4  6.24  3.99  0.72  1.28  1.35      Sham chew          CTL  5.71b  3.78b  6.00b  2.26  0.76  0.36  1.20  0.01          FLX  11.0a  8.37a  12.5a  2.25  0.84  1.29  1.31          FAS  5.34b  6.97a  4.05b  2.65  0.18  1.03  1.19  a,bWithin a column, least squares means lacking common superscripted letters differ (P < 0.05). 1Data represent 96 pregnant sows across 8 replicates, with sow as the experimental unit. 2CTL = standard stall; FLX = flex stall; FAS = free-access stall; ONF = oral–nasal–facial. 3Time periods were divided into six 4-h time periods across 24 h: period 1 (0300 to 0700 h), 2 (0700 to 1100 h), 3 (1100 to 1500 h), 4 (1500 to 1900 h), 5 (1900 to 2300 h), and 6 (2300 to 0300 h). 4SEp = pooled standard error. 5Probability value for the treatment × time period interaction. View Large During time periods 1, 2, and 3, sows kept in FLX were observed sham chewing more often than were sows kept in CTL (TRT × time period, P < 0.01; Table 4). During time period 2, sows in FAS tended (P = 0.06) to sham chew more often than did sows in CTL (Table 4). Main Effects of Treatment Sow-Related Measures. Sows kept in FAS (215 ± 1.6 kg) and FLX (212 ± 1.7 kg) had greater (P < 0.05) mean BW than did sows kept in CTL (209 ± 1.7 kg). Sow mean BF depth was greater (P < 0.001) for those sows kept in FAS (1.88 ± 0.030 cm) than for sows kept in CTL (1.66 ± 0.031 cm). Sows kept in FAS (3.29 ± 0.045) had greater (P < 0.05) mean BCS than did sows kept in CTL (3.09 ± 0.047), whereas sows in FLX (3.14 ± 0.046) had similar BCS to those sows in CTL. Immune and Behavioral Measures. Probability values and means for main effects of stall TRT on sow immune traits are presented in Table 5. Sows kept in FAS had greater (P < 0.05) total numbers of lymphocytes when compared with sows kept in CTL, and sows in CTL tended to have greater (P < 0.10) total number of neutrophils than sows in FLX (Table 5). The percentage of eosinophils was greater (P < 0.001) for sows kept in CTL than for sows kept in FAS, but the percent of eosinophils was similar between sows in FLX and CTL (Table 5). Mitogen ConA-induced lymphocyte proliferation was greater (P < 0.05) for sows in CTL than for sows in FLX and FAS (Table 5). Plasma cortisol was similar for sows in CTL and FAS but was greater (P = 0.05) for sows in FAS when compared with sows in FLX (Table 5). Table 5. Main effects of stall space treatment on endocrine and immune statuses for gestating sows1   Stall treatment3      Immune trait2  CTL  FLX  FAS  SEp4  P-value  Total WBC, 107/mL  2.4  2.2  2.3  0.06  0.07  Lymphocyte, 107/mL  2.6b  2.5b  2.9  0.10  0.02  Neutrophils, 106/mL  5.0  4.7  5.3  0.18  0.09  Lymphocytes, %  52.1  50.2  53.2  1.0  0.12  Neutrophils, %  38.8  39.9  38.9  1.0  0.72  Monocytes, %  3.4  3.8  3.5  0.2  0.53  Eosinophils, %  5.4a  6.0a  4.4b  0.3  0.001  Neutrophil-to-lymphocyte ratio  0.85  0.98  0.93  0.07  0.43  Phagocytosis, %  55.6  56.0  58.2  1.2  0.54  LPS-induced proliferation  1.33  1.39  1.22  0.08  0.28  ConA-induced proliferation  1.76a  1.25b  1.38b  0.13  0.02  NK cell cytotoxicity, % 25:1  43.3  39.8  44.9  5.4  0.79  Chemotaxis, IL-8  59.6  53.1  54.7  6.7  0.78  Chemotaxis, C5a  60.0  50.7  54.1  5.7  0.47  Plasma cortisol, ng/mL  30.2ab  28.7b  33.9a  1.0  0.05    Stall treatment3      Immune trait2  CTL  FLX  FAS  SEp4  P-value  Total WBC, 107/mL  2.4  2.2  2.3  0.06  0.07  Lymphocyte, 107/mL  2.6b  2.5b  2.9  0.10  0.02  Neutrophils, 106/mL  5.0  4.7  5.3  0.18  0.09  Lymphocytes, %  52.1  50.2  53.2  1.0  0.12  Neutrophils, %  38.8  39.9  38.9  1.0  0.72  Monocytes, %  3.4  3.8  3.5  0.2  0.53  Eosinophils, %  5.4a  6.0a  4.4b  0.3  0.001  Neutrophil-to-lymphocyte ratio  0.85  0.98  0.93  0.07  0.43  Phagocytosis, %  55.6  56.0  58.2  1.2  0.54  LPS-induced proliferation  1.33  1.39  1.22  0.08  0.28  ConA-induced proliferation  1.76a  1.25b  1.38b  0.13  0.02  NK cell cytotoxicity, % 25:1  43.3  39.8  44.9  5.4  0.79  Chemotaxis, IL-8  59.6  53.1  54.7  6.7  0.78  Chemotaxis, C5a  60.0  50.7  54.1  5.7  0.47  Plasma cortisol, ng/mL  30.2ab  28.7b  33.9a  1.0  0.05  a–cWithin a row, least squares means lacking common superscripted letters differ (P < 0.05). 1Data represent 96 pregnant sows across 8 replicates, with sow as the experimental unit. 2WBC = white blood cells; LPS = lipopolysaccharide; ConA = concanavalin A; NK = natural killer. 3CTL = standard stall; FLX = flex stall; FAS = free-access stall. 4SEp = pooled standard error. View Large Table 5. Main effects of stall space treatment on endocrine and immune statuses for gestating sows1   Stall treatment3      Immune trait2  CTL  FLX  FAS  SEp4  P-value  Total WBC, 107/mL  2.4  2.2  2.3  0.06  0.07  Lymphocyte, 107/mL  2.6b  2.5b  2.9  0.10  0.02  Neutrophils, 106/mL  5.0  4.7  5.3  0.18  0.09  Lymphocytes, %  52.1  50.2  53.2  1.0  0.12  Neutrophils, %  38.8  39.9  38.9  1.0  0.72  Monocytes, %  3.4  3.8  3.5  0.2  0.53  Eosinophils, %  5.4a  6.0a  4.4b  0.3  0.001  Neutrophil-to-lymphocyte ratio  0.85  0.98  0.93  0.07  0.43  Phagocytosis, %  55.6  56.0  58.2  1.2  0.54  LPS-induced proliferation  1.33  1.39  1.22  0.08  0.28  ConA-induced proliferation  1.76a  1.25b  1.38b  0.13  0.02  NK cell cytotoxicity, % 25:1  43.3  39.8  44.9  5.4  0.79  Chemotaxis, IL-8  59.6  53.1  54.7  6.7  0.78  Chemotaxis, C5a  60.0  50.7  54.1  5.7  0.47  Plasma cortisol, ng/mL  30.2ab  28.7b  33.9a  1.0  0.05    Stall treatment3      Immune trait2  CTL  FLX  FAS  SEp4  P-value  Total WBC, 107/mL  2.4  2.2  2.3  0.06  0.07  Lymphocyte, 107/mL  2.6b  2.5b  2.9  0.10  0.02  Neutrophils, 106/mL  5.0  4.7  5.3  0.18  0.09  Lymphocytes, %  52.1  50.2  53.2  1.0  0.12  Neutrophils, %  38.8  39.9  38.9  1.0  0.72  Monocytes, %  3.4  3.8  3.5  0.2  0.53  Eosinophils, %  5.4a  6.0a  4.4b  0.3  0.001  Neutrophil-to-lymphocyte ratio  0.85  0.98  0.93  0.07  0.43  Phagocytosis, %  55.6  56.0  58.2  1.2  0.54  LPS-induced proliferation  1.33  1.39  1.22  0.08  0.28  ConA-induced proliferation  1.76a  1.25b  1.38b  0.13  0.02  NK cell cytotoxicity, % 25:1  43.3  39.8  44.9  5.4  0.79  Chemotaxis, IL-8  59.6  53.1  54.7  6.7  0.78  Chemotaxis, C5a  60.0  50.7  54.1  5.7  0.47  Plasma cortisol, ng/mL  30.2ab  28.7b  33.9a  1.0  0.05  a–cWithin a row, least squares means lacking common superscripted letters differ (P < 0.05). 1Data represent 96 pregnant sows across 8 replicates, with sow as the experimental unit. 2WBC = white blood cells; LPS = lipopolysaccharide; ConA = concanavalin A; NK = natural killer. 3CTL = standard stall; FLX = flex stall; FAS = free-access stall. 4SEp = pooled standard error. View Large Both duration (P < 0.05) and frequency (P < 0.0001) of sham chew behavior was greater for those sows kept in FLX compared with sows kept in either CTL or FAS (Fig. 3). Mean duration of lay, sit, stand, and ONF behaviors were similar among sows, regardless of TRT. Only frequency of standing behavior was affected by stall TRT; sows kept in FLX (15.8 ± 1.2) stood more often (P < 0.05) than did sows kept in CTL (10.9 ± 1.1) and FAS (11.3 ± 1.1). Figure 3. View largeDownload slide Main effects of stall treatment (CTL = standard stall; FLX = flex stall; FAS = free-access stall) on mean (A) frequency of sham chew behavior (P < 0.0001) and (B) duration of sham chew behavior (P < 0.05) for gestating sows kept in different housing environments. Both frequency and duration for sham chew behavior was greater for sows kept in FLX than for those kept in other stall treatments for each 4-h period. Figure 3. View largeDownload slide Main effects of stall treatment (CTL = standard stall; FLX = flex stall; FAS = free-access stall) on mean (A) frequency of sham chew behavior (P < 0.0001) and (B) duration of sham chew behavior (P < 0.05) for gestating sows kept in different housing environments. Both frequency and duration for sham chew behavior was greater for sows kept in FLX than for those kept in other stall treatments for each 4-h period. DISCUSSION The main focus of this study was to investigate the impact that stall space allowance and access to a small pen area may have on the well-being of gestating sows housed in different stall environments using various welfare metrics. The results indicate that the different stall environments that differed in space allowance and design had positive and negative effects on sow lesion scores, sow- and litter-related traits, and postural and sham chew behaviors. More specifically, increased stall space, especially as pregnancy progressed, resulted in less severe lesion scores, more postural changes, and improved sow- and litter-related traits for sows kept in FLX. Those that gestated in FAS had improved sow- and litter-related traits and spent more time lying but had more severe lesion scores. Higher-parity sows kept in the conventional stalls had more severe lesion scores and reduced litter-related traits. Moreover, compromised welfare measures found among sows in various stall environments may be partly attributed to the specific constraints of each stall system such as restricted stall space and inadequate space allowance in open-pen area or even stall design per se of the FLX (e.g., direction of bars on front gate), which may contribute to differences in sham chew behavior among sows in these stalls. These results also indicate that sow parity and gestational day are additional factors that may exacerbate the effects of restricted stall space or insufficient open-pen space, further compromising sow well-being. One of the major criticisms of the gestation stall is the restricted stall space, which hinders freedom of movement and the sow's ability to perform all normal behaviors while limiting social interactions. Curtis et al. (1989) had previously estimated the dynamic space requirements based on sow BW, body dimensions (width, length, and height), and postural adjustments for dry sows but did not take into consideration the feeding space as part of the floor space; therefore, the dynamic space needs of today's sows is much greater than previously reported. The impacts of stall space allowance has been assessed based on postural behaviors and quantifying the level of injuries that often result from stall confinement of the gestating sow (Boyle et al., 1999; Anil et al., 2002a; Li and Gonyou, 2007). Stall environment per se had minimal to no impact on mean durations of postural behaviors, with the exception that frequencies of lay, sit, and stand behaviors were greater for sows kept in FLX. Increased behavioral frequencies among sows in FLX may be partly reflective of the additional space provided to the sows, which may have made it easier for these sows to stand, sit, and lay. Researchers found that sows kept in 70-cm-wide stalls, which are wider than the average flex stall, made fewer postural changes but sows spent more time lying as stall width increased (Li and Gonyou, 2007), while Barnett et al. (2011) found no effect on frequency of laying. However, we found that adjusting the width of the FLX to accommodate the sow body size at the beginning of gestation, and then again in late gestation so that sows were not touching the sides of the stall, did not affect durations of postural behaviors, only behavioral frequencies. The act of sitting is an unavoidable transition during postural change from lying to standing. Anil et al. (2002b) found that a reduction in stall width slows the act of standing; we found no changes in total time spent lying, sitting, or standing when stall width was increased or restricted. The time it took sows to transition from one posture position to the next was not assessed in this study, so we speculate that it may have taken the same amount of or even less time for sows to transition from one to the other due to the lack of change in durations of postural changes. Also, keeping sows in stalls that are relative to their body size resulted in more frequent transitions from standing to lying and from sitting to standing (Anil et al., 2002b, 2006). Moreover, as pregnancy progresses, stall space becomes more restrictive, sows will make fewer postural changes and spend more time lying due to limited space, and larger sows take longer to transition from one posture to the other (Anil et al., 2002b; McGlone et al., 2004a; Li and Gonyou, 2007). However, stall environment did not affect the time spent lying, even when stall space adjustments were made, which agrees with findings by Li and Gonyou (2007). Stall environment did not impact mean durations and frequencies of various behaviors, but time of day influenced circadian patterns of lying behavior among sows in FAS whereas stand, sit, and sham chew behaviors were affected among sows kept in FLX. Sows in the FAS spent more time lying from 1900 through 2300 h but less time from 2300 through 0300 h. Space use within the open area was not assessed, but it is plausible that the decrease in time sows spent lying may be reflective of sows using the open-pen area in the free-access system. DeDecker (2011) found that heavier-bodied, more dominant sows kept in the FAS system lay more often in the open-pen area as pregnancy progresses. Rioja-Lang et al. (2013) concluded that larger sows use the open-pen area more than smaller sows most likely due to the space restriction within the individual stall. However, it is plausible that sows in this study did not spend more time in the open-pen area due to insufficient space allowance in open-pen area, especially among larger-bodied sows, which may partly explain the lack of behavioral differences found among stall environments. Therefore, if space in the open-pen area is insufficient, sows will spend the majority of their time in the stall instead of the open pen, which is in agreement with findings by DeDecker (2011) and Mack et al. (2014), which both indicated that space and pen design can influence time spent in an open area. It is surprising that the additional stall space in the FLX or in the open-pen area of the FAS affected neither duration nor frequency of ONF behaviors; therefore, sows in all stall environments performed similar levels of ONF behavior across a 24-h period. Research has shown that feed restriction and/or the inability of sows to perform foraging behavior in confined indoor housing systems may contribute to an increase in ONF behaviors. Repetitive chain manipulation and ONF behaviors increased when sows were feed restricted and confined to stalls, thus hindering foraging behavior (Terlouw et al., 1991; Terlouw and Lawrence, 1993; Bergeron et al., 2000). However, these data imply that within each stall environment, sows directed ONF behaviors toward the substrate (e.g., floor, bar, gate) that was available within each environment, implying that gestating sows may be highly motivated to perform ONF behaviors regardless of their housing system because of their strong desire to forage. Interestingly, postural and sham chew behaviors were affected by stall space, time of day, and stall design. Certain patterns of ONF behaviors in anticipation of feeding are common among sows in all housing environments (McGlone et al., 2004a). Stand behavior and sham chew behavior from 0300 until 1500 h (which includes before and after feeding) was much greater among sows in the FLX than among sows in the other stall systems. Dailey and McGlone (1997) concluded that ONF behaviors may be natural pre- and postfeeding appetitive and postconsummatory activities for sows, regardless of housing accommodations. When sows are kept in individual stalls, the availability of environmental features, such as the physical design of the stall, can influence stereotypy behavior. A possible explanation for the increase in both duration and frequency in sham chew behavior may be partly explained by the front gate design. The FLX gate was designed with vertically oriented bars, which hinder the ability of the sow to engage in bar biting; therefore, it is possible that sows redirected pre- and postfeeding oral activity toward sham chew behavior. Some have speculated that oral activity among sows may not always equate to stereotypy behavior or poor welfare but may be indicative of a strong desire to engage in oral behavior that is hindered due to lack of the appropriate substrate; therefore, oral activity is redirected toward the available substrate (Dailey and McGlone, 1997; McGlone et al., 2004b; Salak-Johnson et al., 2007). It is likely that the FLX gate design partly influenced the oral activity and standing behavior displayed by the sows in these stalls because sows were often observed (live and on video) manipulating the one horizontal bar at the top of the gate that was part of the apparatus used to increase the width of the stall. These results, taken together with the visual observations (e.g., manipulation of horizontal bar), may imply that the body size and motivation of the gestating sow may affect the duration and frequencies of various behaviors and that these measures may also be further influenced by other physical attributes within a housing environment. Lesion severity scores are often used to assess the well-being of sows in gestation stalls and group-housing systems. Here, we found that lesions scores were affected not only by stall environment but also by sow parity and gestational day. Lesion scores were less severe among sows that gestated in FLX regardless of gestational day, and lesion severity scores were similar among all parities of sows kept in these stalls, which is most likely due to the increased stall space because higher-parity sows housed in CTL had greater skin lesion scores late in gestation. These results imply that perhaps providing higher-parity sows more stall space or access to an open-pen area as pregnancy progresses can reduce lesion severity scores. However, if space allowance within the open-pen area is insufficient (as it was in this study), this may exacerbate lesion severity scores among higher-parity sows that gestate in a FAS system. The insufficient space in the open-pen area may have hindered the ability of the sows to move freely in and out of the stalls. Boe et al. (2011) reported that for sows to be able to turn around easily, they require a pen width that equals at least 50% of their body length and if pen width is reduced by 60% of sow body length, then lying increases and standing decreases. It may be plausible that the increase in lesion severity scores is due to an increase in aggressive interactions within the open-pen area of the FAS system. However, in this study, we speculate that the high lesion scores early on (at time of mixing) are primarily due to aggressive encounters that occur between sows at the time of mixing, but late in gestation, aggressive encounters are not the cause of higher lesion scores among sows in the FAS system. DeDecker (2011) reported the behavioral observations in the open-pen area among these sows and found that aggressive encounters in the open pen had subsided and did not continue beyond gestational d 45. We speculate that the more severe lesion scores among these sows (late in gestation) is more likely due to insufficient space within the open-pen area, especially among heavier-bodied, higher-parity sows, than aggressive encounters. Both sow BW and parity have been shown to affect lesion severity scores among sows kept in small groups. Salak-Johnson et al. (2007, 2012) reported that heavier-bodied, higher-parity sows housed in small group pens at different floor space allowances had greater lesion scores late in gestation but that these scores were not directly related to an increase in aggressive encounters among the group. In fact, it was more likely that lesion severity scores increase among sows kept in the FAS system and CTL due to space becoming limiting as sows become larger bodied and heavier. Other measures used to assess sow well-being are immune status, performance, and productivity, which, for the most part, were similar among sows kept in different stall housing environments, with the exception of sow BW, BF depth, litter size, and a few immune traits. Sow BW was greater among sows kept in either the FLX or the free-access system. Sows kept in group pens tend to have greater BW than sows in stalls, which may be due to accumulation of more muscle mass during pregnancy due to the increase opportunity to walk (Marchant and Broom, 1996; Salak-Johnson et al., 2007). We did not observe walk behavior; however, it is unlikely that the increased BW among sows in fFAS system is due primarily to walking because only parity 3 and 4 sows were heavier and had deeper BF depth than sows in the other stall environments. Sows kept in the flex stalls had greater mean BW, which may be reflective of larger litters compared with sows kept in the CTL. Moreover, mean BW and smaller litter sizes among higher-parity sows in the conventional stall may also be indicative of the consequences of stall space constraint in terms of sow well-being. These data imply that adequate space allowance within an individual gestation stall and/or in an open-pen area of a FAS system is essential to the well-being of a dry sow, especially larger-bodied sows. Increasing space allowance of an individual gestation stall, especially among higher-parity sows from mid to late gestation, or providing sows access to an open-pen area with adequate space allowance (especially late in gestation) can have positive effects on well-being of gestating sows. Also, it is important to consider other factors that may affect adequate space allowance within a stall system including sow body size, parity, and stage of gestation and, therefore, impact sow well-being. Furthermore, these data imply that it is essential to meet both the static and dynamic space requirements of the dry sow throughout gestation if producers want to improve sow well-being. LITERATURE CITED Anil L. Anil S. Deen J. 2002a. Evaluation of the relationship between injuries and size of gestation stalls relative to size of sows. J. Am. Vet. Med. Assoc.  221: 834– 836. doi: https://doi.org/10.2460/javma.2002.221.834. Google Scholar CrossRef Search ADS   Anil L. Anil S. Deen J. 2002b. Relationship between postural behaviour and gestation stall dimensions in relation to sow size. Appl. Anim. Behav. Sci.  77: 173– 181. doi: https://doi.org/10.1016/S0168-1591(02)00048-5. Google Scholar CrossRef Search ADS   Anil L. Anil S.S. Deen J. Baido S. K. 2006. Cortisol, behavioral responses, and injury scores of sows housed in gestation stalls. J. Swine Health Prod.  14: 196– 201. Barnett J. L. Hemsworth P. H. Butler K. L. Schirmer B. N. Borg S. S. Cronin G. M. 2011. Effects of stall dimensions on the welfare of pregnant sows. Anim. Prod. Sci.  51: 471– 480. doi: https://doi.org/10.1071/EA07124. Google Scholar CrossRef Search ADS   Bergeron R. Bolduc J. Ramonet Y. Meunier-Salaun M. C. Robert S. 2000. Feeding motivation and stereotypies in pregnant sows fed increasing levels of fibre and/or food. Appl. Anim. Behav. Sci.  70: 27– 40. doi: https://doi.org/10.1016/S0168-1591(00)00142-8. Google Scholar CrossRef Search ADS PubMed  Boe K. E. Cronin G. M. Andersen I. L. 2011. Turning around by pregnant sows. Appl. Anim. Behav. Sci.  133: 164– 168. doi: https://doi.org/10.1016/j.applanim.2011.05.002. Google Scholar CrossRef Search ADS   Boyle L. A. Leonard F. C. Lynch P. B. Brophy P. 1999. Prevalence and severity of skin lesions in sows housed individually during production cycle. Ir. Vet. J.  52: 601– 605. Coffey R. D. Parker G. R. Laurent K. M. 1999. Assessing sow body condition.  Cooperative Extension Service, University of Kentucky, Lexington, KY. Curtis S. E. Hurst R. J. Gonyou H. W. Jensen A. H. Muehling A. J. 1989. The physical space requirement of the sow. J. Anim. Sci.  67: 1242– 1248. Google Scholar CrossRef Search ADS   Dailey J. W. McGlone J. J. 1997. Oral/nasal/facial and other behaviors of sows kept individually outdoors on pasture, soil or indoors in gestation crates. Appl. Anim. Behav. Sci.  52: 25– 43. doi: https://doi.org/10.1016/S0168-1591(96)01099-4. Google Scholar CrossRef Search ADS   DeDecker A. E. 2011. Effect of alternative, individual, and group housing systems, and management factors in group pens on the well-being of gestating sows. PhD Diss., Univ. of Illinois at Urbana-Champaign, Urbana, IL. Jolie R. Backstrom L. Olson L. Chase C. 1997. Respiratory and systemic health parameters in pigs raised in a conventional farm or in isolation. Swine Health Prod.  7: 269– 275. Li Y. Z. Gonyou H. W. 2007. Effects of stall width and sow size on behavior of gestating sows. Can. J. Anim. Sci.  87: 129– 138. Google Scholar CrossRef Search ADS   Lumpkin E. A. McGlone J. J. 1992. A 51Cr release assay for the determination of natural killer cell cytotoxicity. J. Nutr. Immunol.  1: 63– 74. doi: https://doi.org/10.1300/J053v01n03_08. Google Scholar CrossRef Search ADS   Mack L. A. Lay D. C.Jr. Eicher S. D. Johnson A. K. Richert B. T. Pajor E. A. 2014. Group space allowance has little effect on sow health, productivity, or welfare in a free-access stall system. J. Anim. Sci.  92: 2554– 2567. doi: https://doi.org/10.2527/jas.2013-7352. Google Scholar CrossRef Search ADS PubMed  Marchant J. N. Broom D. M. 1996. Effects of dry sow housing conditions on muscle weight and bone strength. J. Anim. Sci.  62: 105– 113. doi: https://doi.org/10.1017/S1357729800014387. Google Scholar CrossRef Search ADS   McGlone J. J. Vines B. Rudine A. C. DuBois P. 2004a. The physical size of gestating sows. J. Anim. Sci.  82: 2421– 2427. Google Scholar CrossRef Search ADS   McGlone J. J. von Borell E. H. Dean J. Johnson A. K. Levis D. G. Meunier-Salaun M. Morrow J. Reeves D. Salak-Johnson J. L. Sundberg P. L. 2004b. Review: Compilation of the scientific literature comparing housing systems for gestating sows and gilts using measures of physiology, behavior, performance, and health. Prof. Anim. Sci.  20: 105– 117. Moustsen V. A. Lahrmann H. P. D'Eath R. B. 2011. Relationship between size and age of modern hyper-prolific crossbred sows. Livestock Sci.  141: 272– 275. Google Scholar CrossRef Search ADS   Niekamp S. R. Sutherland M. A. Dahl G. E. Salak-Johnson J. L. 2007. Immune responses of piglets to weaning stress: Impacts of photoperiod. J. Anim. Sci.  85: 93– 100. doi: https://doi.org/10.2527/jas.2007-153. Google Scholar CrossRef Search ADS PubMed  NRC 1998. Nutritional requirements of swine.  Natl. Acad. Press, Washington, DC. Rioja-Lang F. C. Hayne S. M. Gonyou H. W. 2013. The effect of pen design on free space utilization of sows group housed in gestation pens equipped with free access stalls. Appl. Anim. Behav. Sci.  148: 93– 98. doi: https://doi.org/10.1016/j.applanim.2013.07.002. Google Scholar CrossRef Search ADS   Salak-Johnson J. L. DeDecker A. E. Horsman M. J. Rodriguez-Zas S. L. 2012. Space allowance for gestating sows in pens: Behavior and immunity. J. Anim. Sci.  90: 3232– 3242. doi: https://doi.org/10.2527/jas.2011-4531. Google Scholar CrossRef Search ADS PubMed  Salak-Johnson J. L. Niekamp S. R. Rodriguez-Zas S. L. Ellis M. Curtis S. E. 2007. Space allowance for dry pregnant sows in pens: Body condition, skin lesions, and performance. J. Anim. Sci.  85: 1758– 1769. doi: https://doi.org/10.2527/jas.2006-510. Google Scholar CrossRef Search ADS PubMed  Sutherland M. A. Rodriguez-Zas S. L. Ellis M. Salak-Johnson J. L. 2005. Breed and age affect baseline immune traits, cortisol, and performance in growing pigs. J. Anim. Sci.  83: 2087– 2095. Google Scholar CrossRef Search ADS PubMed  Terlouw E. M. C. Lawrence A. B. 1993. Long-term effects of food allowance and housing on development of stereotypies in pigs. Appl. Anim. Behav. Sci.  38: 103– 126. doi: https://doi.org/10.1016/0168-1591(93)90060-3. Google Scholar CrossRef Search ADS   Terlouw E. M. Lawrence A. B. Illius A. W. 1991. Influences of feeding level and physical restriction on development of stereotypies in sows. Anim. Behav.  42: 981– 991. doi: https://doi.org/10.1016/S0003-3472(05)80151-4. Google Scholar CrossRef Search ADS   Whittemore C.T. 1994. Causes and consequences of change in the mature size of the domestic pig. Outlook Agric.  23: 55– 59. Google Scholar CrossRef Search ADS   Zverina L. R. 2011. A pilot study on productivity and behavior of the gestating sow housed in a flex stall. MS Thesis, Univ. of Illinois at Urbana-Champaign, Urbana, IL. Footnotes 1 The research herein was supported by the National Pork Board. The authors gratefully acknowledge Tom Crenshaw and John Kane for development and fabrication of the Flex stall and Chore-Time and Laake for supplying free-access stall accommodation. American Society of Animal Science
TI - Wider stall space affects behavior, lesion scores, and productivity of gestating sows
JF - Journal of Animal Science
DO - 10.2527/jas.2015-9017
DA - 2015-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/wider-stall-space-affects-behavior-lesion-scores-and-productivity-of-O1EADQM2yP
SP - 5006
EP - 5017
VL - 93
IS - 10
DP - DeepDyve
ER -