Applied Animal Behaviour Science 246 (2022) 105527 Contents lists available at ScienceDirect Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim Welfare and performance of sows and piglets in farrowing pens with temporary crating system on a Spanish commercial farm Heng-Lun Ko a, *, Déborah Temple a, Janni Hales b, Xavier Manteca a, Pol Llonch a a Department of Animal and Food Science, School of Veterinary Science, Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Barcelona, Spain b SKIOLD JYDEN A/S, Idomvej 2, 7570 Vemb, Denmark A R T I C L E I N F O A B S T R A C T Keywords: The study aimed to compare the welfare and performance of sows and piglets in three different farrowing sys­ Crushing tems (farrowing crate (FC) and farrowing pen with temporary crating (TC): SWAP and JLF15). One batch of Farrowing system crossbred Duroc was followed in every season. There were 18 sows (183 piglets) in FC, 23 sows (243 piglets) in Maternal behavior SWAP, and 23 sows (237 piglets) in JLF15. Farrowing day was Day (D) 0 and weaning occurred on D24. Crating Mother-young interaction Stress biomarker period was from entry to weaning in FC, and from 1-day pre-expected farrowing day to D3 in TC. Social in­ Temporary crating teractions between littermates, between sows and piglets, and exploration by sows and piglets were observed on D2, D4, D12, and D23. Video recordings of crushing events which led to piglets’ death were studied. Piglets were weighed on D3 and D19, and foreleg abrasions were assessed on D19. Sow saliva samples were collected on 1 day before and after temporary confinement, D2, D4, D12, and D23 to evaluate cortisol (CORT) and chromogranin A (CgA) levels. TC piglets initiated naso-naso contacts with sows more frequently than FC on D2 (P ≤ 0.01). TC sows interacted with piglets more frequently than FC on D12 (P ≤ 0.05). TC sows explored the pens more frequently than FC on D4 (P ≤ 0.05). Crushing rate (i.e. number of piglets per sow) of SWAP (1.2 ± 0.3) was higher than those of JLF15 (0.6 ± 0.2) and FC (0.3 ± 0.1) (P ≤ 0.02), and crushing rate of JLF15 was higher than that of FC (P < 0.0001). Autumn batch showed the lowest crushing rate amongst different batches (P < 0.001). Number of crushing incidents was similar when TC sows were crated or loose (P = 0.54). The percentage of occasions when sows used a support from the pen when changing posture but still crushed the piglets was higher in FC than in TC (P = 0.05). No difference of growth and foreleg abrasion in piglets were found between systems. CORT in SWAP peaked on D2 (P = 0.02), and CgA in JLF15 peaked on D4 and remained elevated on D12 (P ≤ 0.05). CORT and CgA in FC remained similar during lactation. Our results suggested that TC facilitated mother- young interactions; TC showed a higher crushing rate; sows changed their posture differently between systems and batches; the practice of temporary crating did not alter the level of stress biomarkers in sows. 1. Introduction European Union in particular, it is estimated that 95% of the pig farms are using FC (Johnson and Marchant-Forde, 2009). However, while it Farrowing crates (FC) were first developed in the 1950 s and became effectively prevents piglet’s death from crushing (Nicolaisen et al., popular since then (Mellor et al., 2009). This space-saving design of the 2019), FC presents some welfare concerns for the sows because it limits farrowing pen allows more sows to farrow per unit, permits easy in­ the performance of some natural behaviors such as: body movement, spection and safe intervention on sows and piglets by farm staff (Chidgey nest building and maternal behavior (i.e. interaction with the piglets) et al., 2015), and most importantly, it restricts sows’ posture changes (Baxter et al., 2012). with the purpose of reducing piglet crushing (Baxter et al., 2018). As the In natural conditions, the sow seeks an enclosed site to build a pro­ loss of piglets is an economic and welfare concern (Chidgey et al., 2015), tective nest prior to giving birth (Jensen, 1986). Nest-building behavior this highly cost-effective system has therefore been widely accepted and is still observed in domestic pigs (Jensen, 1986), even when the sow is installed on pig farms all over the world (Hales et al., 2016). Within the confined in FC, with redirected behavior towards the pen fixtures like * Corresponding author. E-mail addresses: henglun.ko@uab.cat (H.-L. Ko), deborah.temple@uab.cat (D. Temple), jhp@skiold.com (J. Hales), xavier.manteca@uab.cat (X. Manteca), pol. llonch@uab.cat (P. Llonch). https://doi.org/10.1016/j.applanim.2021.105527 Received 11 June 2021; Received in revised form 25 November 2021; Accepted 26 November 2021 Available online 29 November 2021 0168-1591/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 floor, bars, and drinker, which is associated with nest-building (Baxter et al., 2018). Research has found that the structure of FC prevents sows from fully expressing nest-building behavior, as it narrows sows’ movements down to standing, sitting, lying, and rolling (Chidgey et al., 2016), which later may reflect on their poor reproductive performance, although not al­ ways (Hales et al., 2015; Hansen et al., 2017; Nowland et al., 2019; Lohmeier et al., 2020). Sows in FCs were found to have a longer far­ rowing duration (93 min longer in Oliviero et al., 2008; 77.6 min longer in Gu et al., 2011) and higher stillborn rates (Oliviero et al., 2010; Gu et al., 2011), compared with loose pen sows. On the other hand, satis­ factory nest-building behavior has been associated with lower risk of crushing (Pedersen et al., 2003; Andersen et al., 2005), and greater suckling success for piglets due to an increased secretion of oxytocin by sows (Yun et al., 2013). The welfare concerns over farrowing sows have resulted in various designs of alternative farrowing systems in the past decades, farrowing pen with temporary crating system (TC) being one example. With the aim of improving the welfare of both sows and pig­ lets, TC allows the sows to be loose during lactation, except for a few days peripartum to limit the sows’ most dangerous movements to reduce piglet mortality (Moustsen et al., 2013; Hales et al., 2015), although it is difficult to reach the same level of mortality as in FC (Chidgey et al., 2015). Although TC has started to be implemented in some European countries, experience with these systems is still very limited. There are currently many commercially available TC in the market, but most of the studies compared the farrowing systems between FC and TC (e.g. Chidgey et al., 2015; Hales et al., 2015), or FC and loose pen (e.g. Hales et al., 2014), rather than between different designs of TC. The objective Fig. 1. Distribution of the farrowing systems tested in the present study. There of the present study was to assess the effect of farrowing systems, were five pens of conventional system with farrowing crates (FC) in one far­ including one FC and two types of TC, on the welfare and performance of rowing unit, and six SWAP pens and six JLF15 pens which are the temporary sows and piglets on a commercial farm in Spain. crating systems in another farrowing unit. The arrows indicate the entrances to the farrowing units. The pig drawings indicate the location of the sows when 2. Materials and methods being crated, except those on the top pens of SWAP and JLF15. The top pens of SWAP and JLF15 indicate the area available to the loose sows. The squares which confine the sows indicate the farrowing crates (the solid lines are the 2.1. Housing and experimental design fixed sides, and the round-dotted lines are the swing-sides which are adjustable) (crating period for FC: from entry to weaning; for SWAP and JLF15: from 1-day The study was conducted on a farrow-to-finish commercial farm pre-expected farrowing date to 3 days after farrowing). The orange circles (Gerfam SL; Girona, Spain). One week before the expected farrowing indicate the lamps for piglets. The grey trapezoids in SWAP and JLF15 pens day, sows were moved from the gestation unit to one of the three far­ indicate the creep areas for piglets. The squares with the zig zag pattern filling rowing systems tested in the present study: one conventional system in FC pens indicate the heating mats for piglets. The squares with the narrow with a FC (FC) and two commercially available TC: Sow Welfare and vertical stripe pattern filling in SWAP and JLF15 pens indicate the straw racks. Piglet protection pen (SWAP) and JLF15 (both produced by Jyden; The bracket shapes in SWAP and JLF15 pens indicate the piglet protection rails. Sæby, Denmark). There were five FC pens in one farrowing unit, and six The squares with the dark vertical stripe pattern filling indicate the metal- SWAP pens and six JLF15 pens in another farrowing unit. Fig. 1 illus­ barred gates in SWAP pens. The sloping walls in SWAP pens which are not trates the pen distribution and the key features of each farrowing sys­ indicated in the figure are installed at the fixed sides of the crates. Technical details of the pen and the creep sizes for each farrowing system are presented tem. The floor of both farrowing units was adapted to deep slurry system in Table 1. which is the commercial conditions in Spain. SWAP and JFL15 were equipped with plastic solid flooring for the creep area and the straw rack area, cast-iron slat flooring for the crating area, and plastic slat flooring 3.2 ± 0.5; SWAP: 3.2 ± 0.4; JLF15: 3.3 ± 0.3). The study period started for the rest of the pen. FC had cast-iron slat flooring for the sows, plastic from sow entry and ended at weaning. The day of the farrowing of each slat flooring for the rest of the pen, and a heat mat for piglets. Two heat pen was considered as Day (D) 0. Crating period of FC sows was from mats were installed at the floor of the creep area of SWAP and JLF15, entry to weaning while that of SWAP and JLF15 was from 1-day pre- and a lid at the top to retain the heat and to facilitate daily inspection of expected farrowing date to 3 days after farrowing (i.e. the crate the piglets. A lock-in practice of the piglets is possible in SWAP and opened on D3). Table 1 summarizes the technical details of each far­ JLF15 by closing the creep area with a gate when necessary. Above the rowing system. heat mat of FC and the creep area of SWAP and JLF15, lamps (150 W) as an additional heat source for piglets were installed. The straw racks in 2.2. Animals and management SWAP and JLF15 were refilled with hay twice a day (08:00 and 18:00 h) from sow entry to weaning. The temperature in the farrowing units was Management routines and handling of sows and piglets were con­ kept constant at ~20 ◦ C, and the light was on from 07:00–18:00 h every ducted in accordance with the routine husbandry of the farm. Shredded day. newspaper was provided at the end of the crate and the creep area in all Four batches of crossbred Duroc sows (n = 68) were randomly pens before farrowing. Every hour, newborn piglets were inspected for allocated to the three farrowing systems (FC, SWAP and JLF15) in all their body conditions and the colostrum intake, and creep areas were four seasons between 2018 and 2019: autumn batch (November), winter monitored for the cleanliness and the temperature. Litter size was batch (February), spring batch (May) and summer batch (July). Parity of standardized at 10 piglets by cross-fostering within 72 h after birth. the sows in each batch was balanced between the farrowing systems (FC: Piglets received an iron injection and were teeth-clipped following 2 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 Table 1 Labiana; Terrassa, Spain) and creep feed (Nuscience; Ghent, Belgium) Technical details of each farrowing system tested in the study, including one from D10 to weaning. Both sows and piglets had free access to water conventional farrowing crate (FC) and two commercially available farrowing during the study period. pens with temporary crating (SWAP and JLF15). A total of 674 piglets, identified individually by a numeric ear tag, FC SWAP JLF15 were included in the study. Piglets were weaned at 24 days of age and Farrowing Farrowing Farrowing pen with Farrowing pen with were moved to another unit of the farm equipped with conditioned in­ system crate temporary crating temporary crating frastructures for young weaners. Crating period From entry to From 1 day before From 1 day before weaning expected farrowing expected farrowing date to 3 days date to 3 days 2.3. Direct behavioral observations postpartum (D3a) postpartum (D3a) Number of 5 6 6 Social interactions between sows and piglets, play in piglets, and pens per exploration by piglets and sows, were recorded through direct obser­ batch vations by one observer. Behavioral observation took place on D2, D4, Pen size (m x 2.65 × 1.50 3.00 × 2.00 2.40 × 2.40 m) D12 and D23. Each observation day comprised six sessions, three in the Crate size 2.2 × 0.6 2.35 × 0.86 2.40 × 0.62 morning (from 10:00–13:30 h) and three in the afternoon (from (including 14:00–17:30 h). All pens were observed in these six sessions. Behaviors the feed were recorded by using 30-second scan-sampling for 3 min per pen in trough, m x m) each session (i.e. 30-second scan-sampling for three times per pen per Creep area None, but a 0.9 0.9 session). The observation order was rotated between the farrowing (m2) heating mat of systems until all the pens were observed. Behavioral categories are 0.4 described in Table 2. Straw rack (cm None Yes, 50 × 50×21 Yes, 45 × 35×11 Intra-observer reliability was calculated based on three pre-recorded x cm x cm) (square shape) (half circle shape) Piglet None Two protection rails One protection rail 3-minute videos of each farrowing system (i.e. nine 3-minute videos in protection (130 and 50 cm) and (130 cm) total), and carried out seven repetitions of the video behavioral obser­ features from one sloping wall vation at different points of the study. crushing (158 ×50 cm2) a The day of the farrowing was considered as Day (D) 0. Crates in SWAP and 2.4. Saliva collection and stress biomarker analysis in sows JLF15 were opened on D3 Saliva samples were collected by introducing the cotton swabs, veterinary recommendation on D3 before the SWAP and JLF15 sows provided in the Salivette® tubes (Sarstedt, Aktiengesellschaft & Co., were set loose. Treatments and manual interventions during farrowing Nümbrecht, Germany), into the sows’ mouths for 1 min. Six sampling followed the usual routine of the farm and were performed by the same points were determined: 1-day before temporary confinement (i.e. of the person. When the time interval between the birth of two piglets excee­ SWAP and JLF15 sows, FC sows remained crated), 1-day after temporary ded 1 h and the cervical canal was dilated, 1 mL of oxytocin (Hormo­ confinement (i.e. of the SWAP and JLF15 sows, FC sows remained nipira, HipraSA; Girona, Spain) was injected. crated), D2 (i.e. 1-day before opening the crate of the SWAP and JLF15 Lactating sows were fed twice a day (07:00 and 18:00 h). Piglets sows, FC sows remained crated), D4 (i.e. 1-day after opening the crate of were supplemented with complementary liquid feed (Re-hydralab, the SWAP and JLF15 sows, FC sows remained crated), D12 and D23 Table 2 Behavior categories recorded through direct observations using scan sampling. Category Subject Behavior Description Social Piglet towards Social behavior (SB) Piglet performing any physical contact, including positive and negative, with one or more piglets. Fighting, interaction piglet(s) a chain of agonistic interactions by at least two individuals; the number of the event is recorded as the number of the individuals involved in the event. Positive and negative social interactions were pooled into one category due to the low number of events in each category observed in the pilot study. Piglet towards Naso-naso contact (NNC) Snout of the piglet approaching or gently touching the snout of its mother or the neighbora sow. sow Sow contact (SC) Piglet performing any physical contact with its mother or the neighbora sow, such as nudging, chewing, climbing on another individual or huddling. Any behavior directed to the sow’s snout or udder was excluded. Resting with sow contact Number of piglets resting next to the sow or approaching towards the sow to rest during the behavioral (RSC) observation. Sow towards Mother-young interaction Sow performing any physical contact with minimal or moderate force towards her piglet(s) or her piglet(s) (MYI) neighbor’sa piglet(s) such as naso-naso contact, sniffing or nudging. Nursing is excluded. Sow towards Mother-mother interaction Sow performing any social interaction or physical contact with the neighborb sow, such as naso-naso sow (MMI) contact, sniffing, nudging or aggression. MMI is not possible in the conventional farrowing crates (FC). Non-social Piglet Locomotor/object play and Piglet performing any locomotor play behaviors including scampering, pivoting, head tossing, flopping, interaction exploration (PPE) hopping, rolling or gamboling (seeMartin et al., 2015 for each definition); object play and exploratory behaviors including sniffing or manipulating the nest-building materials, newspaper, pen facilities or other items; piglet’s feed and water are excluded. Locomotor/object play and exploration were pooled into one category due to the low number of events in each category observed in the pilot study. Sow Exploration (SEB) Sow performing exploratory behavior including sniffing or manipulating the nest-building materials (i.e. hay), newspaper, pen facilities or other items; sow’s feed and water are excluded. Sow performing exploratory behavior continuously for 5 s would count as one event. a Metal-barred gates were installed in SWAP pens, so SWAP sows and piglets of the adjacent pens could interact with each other through the gates. On the other hand, JLF15 piglets could lift their upper part of the bodies through one of the piglet protection rails to interact with the sow from the adjacent pen. b In both SWAP and JLF15 pens, sows of the adjacent pens could interact with each other when they were loose, as both types were installed with low solid walls of 90 cm (SWAP pens) or 50 cm (JLF15 pens) and two or three horizontal metal bars above the walls. 3 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 between 09:00 and 10:00. Saliva samples of the SWAP and JLF15 sows 2.7.2. Direct behavioral observations were obtained by introducing a long stick with the cotton swab attached Behavioral data collected in six sessions on each observation day on a clamp, without the sampler entering their pens. Saliva samples of were summed up for each behavioral category. Proportion of each the FC sows were obtained by the same technique (i.e. cotton swab on a behavior in each pen was calculated by dividing the amount of each clamp) but without the long stick and without entering their pens. All behavior in a pen by the total amount of sample points in one day. Each the sows were trained to be accustomed to the sampling technique behavior category was expressed as percentage. RSC for each observa­ before the first collection commenced. Samples were immediately tion day was obtained by calculating the average of six RSC sample centrifuged (Heraeus™ Labofuge™ 200 Centrifuge, Thermo Fisher Sci­ points on each day. SB and PPE were normally distributed so were entific GmbH, Dreieich, Germany) for 10 min at 3000 rpm and stored at analyzed by linear mixed models (LMM). NNC, SC and MYI were log − 20 ◦ C until analysis. (1 +x) transformed and analyzed by LMMs. 84% of MMI and 45% of SEB Cortisol (CORT) and chromogranin A (CgA) were selected as salivary were 0 s, so data were changed to the value of either 1 or 0 (i.e. Yes or stress biomarkers for measuring the crating stress in sows. CORT was No) and analyzed by general linear mixed models (GLMM) with a detected by an automated chemiluminescence immunoassay (Immulite binomial distribution. RSC was analyzed by a GLMM with a Poisson 1000 cortisol, Siemens Medical Solutions Diagnostics, Los Angeles, CA, distribution. All models had the behavior as the response variable, far­ USA) (as validated in Escribano et al., 2012). CgA was detected by rowing system, day and their interaction as the fixed effects, litter size as time-resolved immunofluorometry assays (TR-IFMA) (as validated in the covariate, and batch and pen as the random effects. Escribano et al., 2013). 2.7.3. Concentration of salivary stress biomarkers in sows 2.5. Video recordings of the crushing events Concentration of CORT and CgA were log transformed to be fitted into LMMs: concentration of CORT or CgA as the response variable, Crushing events which led to the death of the piglets were confirmed farrowing system, sampling day and their interaction as the fixed effects, by surveillance cameras (resolution: 1920 ×1080, 30 FPS) (IMX291, basal level (samples collected on 1-day pre-expected farrowing day) as Megapixel Starvis, Sony). Surveillance cameras (n = 9) were installed on the covariate, and batch and sow as the random effects. the ceiling of the farrowing units to monitor the sows and the piglets during the study, with three cameras covering all the pens in each far­ 2.7.4. Number of crushed piglets per sow and video recordings of the rowing system. Crushing events were analyzed through several param­ crushing events eters: ‘batch,’ ‘parity,’ ‘crushing time,’ ‘sow was crated/loose,’ ‘crushing Number of crushed piglets per sow was analyzed by a general linear day,’ ‘body part of the sow,’ ‘posture change: from posture 1 to posture 2 model (GLM) with a Poisson distribution: number of crushed piglets per (postures include stand, sternal/ventral recumbency, sit, and lateral sow as the response variable, farrowing system, batch, and parity (pri­ recumbency),’ and ‘Did the sow use the pen fixture as an aid to change miparous vs. multiparous) as the fixed effects, and litter size as the her posture? (yes/no)’. Detailed description for each parameter can be covariate. found in Table 6. As for the video recordings of the crushing events, two parameters used to analyze the crushing events, ‘batch’ and ‘parity’, were analyzed by the above-mentioned GLM. For the rest of the parameters (see 2.6. Weighing and foreleg abrasion assessment in piglets Table 3), they were analyzed with several chi-square tests separately to compare the percentage of occurrence between farrowing systems. The After cross-fostering was completed, piglets were ear-tagged and parameter, ‘crated or loose’, was compared only between SWAP and weighed individually on D3. To calculate the average daily gain (ADG) JLF15. in the pre-weaning period, piglets were weighed individually again on D19. On D3, piglets in SWAP and JLF15 were weighed first before their 2.7.5. Growth performance and foreleg abrasion in piglets sows were set loose. Sows in SWAP and JLF15 were crated temporarily When using piglet as the experimental unit, body weight on D3 again on D19 during the weighing of their piglets. Additionally, while (BW3), D19 (BW19), and ADG3–19 in the pre-weaning period were weighing piglets on D19, skin abrasion on the forelegs of the piglets was analyzed by LMMs: BW3, BW19, or ADG3–19 as the response variable, examined on a Yes (1) / No (0) basis (Johansen et al., 2004): “0” indi­ farrowing system, sex and their interaction as the fixed effects, batch and cated no skin abrasion on any forelegs of the piglet, and “1” indicated at sow as the random effects, and BW3 as the covariate (except for the BW3 least one skin abrasion with a minimum size of 1 cm was observed on model). On the other hand, when using litter as the experimental unit, one of the forelegs of the piglet. Table 3 2.7. Statistical analysis Crating period, equalized litter size and selected reproductive parameters of sows by three farrowing systems: the conventional farrowing crate (FC) and two Data were analyzed in RStudio version 1.2.5033 (R Foundation, commercially available farrowing pens with temporary crating (SWAP and Austria). The farrowing pen (i.e. sow or litter) was the experimental unit JLF15). for reproductive performance, direct behavioral observations, parame­ FC SWAP JLF15 P-value ters used for the video recordings of the crushing events, and salivary stress biomarkers. The piglet was the experimental unit for foreleg Number of sows 18 23 23 – Crating period (number of 31.8 ± 0.5 6.0 ± 0.4 5.3 ± 0.3 < 0.0001 abrasion. As for piglet growth performance, we ran two separate ana­ days)a lyses using the farrowing pen or the piglet as the experimental unit. Number of total piglets 11.2 ± 0.6 11.7 ± 0.5 12.4 ± 0.6 0.29 Statistical significance was set at P ≤ 0.05, and a tendency was consid­ born per litter ered when 0.05 < P ≤ 0.10. Data are presented as means with standard Number of piglets born 10.6 ± 0.6 11.0 ± 0.5 11.3 ± 0.6 0.69 alive per litter error ( ± SE). Number of stillborn piglets 0.6 ± 0.2 0.6 ± 0.2 0.8 ± 0.2 0.27 per litter 2.7.1. Descriptive data Equalized litter sizeb 10.4 ± 0.5 10.3 ± 0.2 10.2 ± 0.3 0.83 Number of total born/born alive/stillborn/weaned piglets, crating Number of piglets weaned 9.6 ± 0.4 9.0 ± 0.3 9.3 ± 0.3 0.34 period and equalized litter size between farrowing systems were per litter analyzed with Kruskal-Wallis tests: the above-mentioned variables as a Crating period was not different between SWAP and JLF15 (P = 0.28). the response variable, and farrowing system as the fixed effect. b Establishment of the litter size (within 72 h after birth) after cross-fostering. 4 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 the statistical models were the same as mentioned above except JLF15. including sex and the interaction between farrowing system and sex as the fixed effects. Foreleg abrasion was analyzed by a GLMM with a 3.1. Direct behavioral observations binomial distribution: foreleg abrasion (0/1) as the response variable, farrowing system and sex as the fixed effects, batch and sow as the Intra-observer reliability test was considered acceptable (Cronbach’s random effects. α = 0.98). Percentages of exploration and social behaviors of piglets and sows in the three farrowing systems during lactation are presented in 3. Results Table 4. FC piglets performed more SB than SWAP and JLF15 piglets at an One multiparous FC sow in the summer batch was excluded from the early age (D2 and D4), and this higher amount of social interaction study due to savaging of her newborn piglets, which led to low number between FC piglets tended to continue until weaning. In contrast, SWAP of live piglets. One multiparous FC sow in the summer batch and one and JLF15 piglets performed more NNC than FC piglets on D2, and multiparous JLF15 sow in the spring batch were also excluded from the SWAP and JLF15 sows performed more MYI than FC sows on D12. study due to 1-week delay of farrowing. One multiparous SWAP sow in Moreover, JLF15 and SWAP piglets tended to perform more SC than FC the summer batch was considered as an extreme outlier due to excessive piglets on D2 and D4, respectively, and JLF15 sows tended to perform crushing incidents, which led to eight piglets being crushed, and more MYI than FC sows on D4. Our results suggested that in general, therefore was also excluded from the study. there was a higher amount of social interaction between sows and pig­ Data of behavior, body weight and foreleg abrasion were collected lets in TC during lactation, compared to FC. On D4, 1 day after opening only from the piglets with ear-tags (i.e. the live piglets after the estab­ the crate in TC, SWAP and JLF15 sows performed more SEB than FC lishment of the litter size). Data of number of total born/born alive/ sows. stillborn/crushed/weaned piglets were collected from all the piglets, Overall, SB was negatively and moderately correlated with NNC including those died before the establishment of the litter size. (r = − 0.36), SC (r = − 0.34), and MYI (r = − 0.41) (all P < 0.0001), In the end, there were 18 sows (three primiparous and 15 multipa­ indicating that the more social interactions between littermates, the less rous) (and their 183 piglets) in FC, 23 sows (four primiparous and 19 social interactions between sows and piglets, and vice versa. multiparous) (and their 243 piglets) in SWAP, and 23 sows (two pri­ miparous and 21 multiparous) (and their 237 piglets) in JLF15 in the 3.2. Concentration of salivary stress biomarkers in sows study. Table 3 summarizes the crating period, the litter size after cross- fostering and selected reproductive parameters of sows in FC, SWAP and Concentrations of CORT and CgA during lactation are presented in Table 4 Percentage ((the amount of each behavior/total amount of sample points)*100%) of exploration and social interactions of piglets and sows in three farrowing systems during the lactation period (FC: conventional farrowing crate; SWAP and JLF15: farrowing pens with temporary crating). RSC is presented in number of piglets. Crates in SWAP and JLF15 were opened on Day 3. Social interaction Non-social interaction Piglet towards piglet(s) Piglet towards sow Sow towards piglet(s) Sow towards sow Piglet Sow SBa NNCb SCb RSCc MYIb MMId PPEa SEBd Day 2 FC 40.2 ± 4.3x 3.6 ± 0.7x 11.9 ± 2.4+ 1.6 ± 0.3 4.7 ± 1.3 - 30.9 ± 3.4 3.2 ± 1.3 SWAP 29.9 ± 3.3y 9.6 ± 1.6y 15.8 ± 2.5+* 1.0 ± 0.2 8.8 ± 2.2 - 33.3 ± 2.3 2.6 ± 1.3a JLF15 30.7 ± 3.6y 10.5 ± 2.8y 20.3 ± 2.7* 1.3 ± 0.2 5.6 ± 1.3 - 25.9 ± 4.0a 7.0 ± 3.2 Day 4 FC 41.6 ± 3.4x 4.8 ± 1.2 12.7 ± 2.1+ 1.1 ± 0.2 4.7 ± 1.2+ - 33.4 ± 3.7 2.8 ± 1.5x SWAP 23.0 ± 2.4y 7.5 ± 1.2 21.5 ± 3.8* 0.9 ± 0.2 9.2 ± 1.7+* 0.6 ± 0.3 28.2 ± 2.4 9.9 ± 2.7y,b JLF15 31.5 ± 2.7y 9.0 ± 1.9 14.0 ± 2.1+* 0.9 ± 0.1 10.3 ± 2.9* 0.4 ± 0.4 28.2 ± 2.7a 6.6 ± 1.7y Day 12 (mid-lactation) FC 35.1 ± 4.0+ 4.6 ± 1.8 16.9 ± 3.2 1.0 ± 0.2 2.4 ± 0.8x - 29.9 ± 3.7 5.6 ± 4.1 SWAP 26.1 ± 1.9* 6.8 ± 0.8 20.4 ± 3.2 0.7 ± 0.2 8.2 ± 1.2y 0.6 ± 0.2 32.4 ± 2.2 5.6 ± 1.3b JLF15 26.9 ± 1.6* 8.4 ± 1.3 19.8 ± 1.9 1.0 ± 0.1 7.4 ± 1.3y 0.4 ± 0.2 32.9 ± 2.8ab 4.2 ± 1.0 Day 23 (late lactation) FC 33.0 ± 2.3+ 3.6 ± 0.8 17.5 ± 2.5 0.9 ± 0.2 3.1 ± 0.7 - 39.6 ± 1.4 3.2 ± 1.2 SWAP 24.3 ± 1.8* 5.6 ± 0.6 19.6 ± 2.7 1.1 ± 0.2 7.3 ± 1.0 0.2 ± 0.1 35.5 ± 1.8 7.4 ± 2.1ab JLF15 23.7 ± 1.6* 5.6 ± 1.0 19.3 ± 2.6 1.3 ± 0.2 6.3 ± 1.1 0.2 ± 0.1 39.2 ± 2.1b 5.8 ± 1.5 Global P-value Farrowing system < 0.0001 < 0.0001 0.04 0.87 0.0002 0.16 0.49 0.006 Day 0.06 0.17 0.28 0.20 0.28 0.18 0.003 0.001 System x Day 0.50 0.79 0.40 0.78 0.80 0.61 0.26 0.48 Behaviors initiated by the piglets: SB = Social interactions between piglets. / NNC = Piglet initiated naso-naso contact with the sow. / SC = Piglet initiated physical contact (except the snout and the udder) with the sow. / RSC = Piglet resting in physical contact with the sow. / PPE = Piglet locomotor or object play, and exploration of the pen. Behaviors initiated by the sow: MYI = Sow initiated physical contact with the piglet. / MMI = Mother-mother interactions. / SEB = Sow exploration of the pen. Values with a different letter superscript are significantly different from each other (P ≤ 0.05): x, y indicate difference between farrowing systems in the same behavior category on the same day; a, b indicate difference between days in the same behavior category and farrowing system. Values with a different symbol superscript (+, *) correspond to a tendency of difference between farrowing systems in the same behavior category on the same day (0.05 < P ≤ 0.10). a SB and PPE were normally distributed, so they were analyzed by linear mixed models. b NNC, SC and MYI were log(1+x) transformed and analyzed by linear mixed models. c RSC was analyzed by a general linear mixed model with a Poisson distribution. d Values of MMI and SEB were changed to 1/0 (i.e. Yes/No) and analyzed by general linear mixed models with a binomial distribution. 5 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 Table 5 Concentration of salivary cortisol (CORT) (µg/dL) and chromogranin A (CgA) (µg/mL) of sows in the three farrowing systems on different sampling days during the lactation period (FC: conventional farrowing crate; SWAP and JLF15: farrowing pens with temporary crating). Crates in SWAP and JLF15 were opened on Day 3. -1D cratinga + 1D cratingb Day 2c Day 4d Day 12 Day 23 Global P-values (mid-lactation) (late lactation) Farrowing system Day System x Day CORT, µg/dL FC 0.48 ± 0.14 0.49 ± 0.15 0.35±0.09x 0.38 ± 0.05 0.66 ± 0.19 0.69 ± 0.29 0.008 SWAP 0.68±0.16ab 0.48±0.11a 1.27±0.37y,b 0.55±0.08ab 0.72±0.11ab 0.84±0.18ab < 0.0001 JLF15 0.56 ± 0.11 0.52 ± 0.09 0.64 ± 0.22x 1.10 ± 0.33 0.75 ± 0.14 0.67 ± 0.11 0.19 CgA, µg/mL FC 0.65 ± 0.23 0.73 ± 0.30 0.52 ± 0.12 0.95 ± 0.33 1.18 ± 0.27 0.69 ± 0.18 0.51 SWAP 0.81 ± 0.18 0.62 ± 0.14 0.68 ± 0.18 1.01 ± 0.24 0.81 ± 0.18 0.97 ± 0.23 < 0.0001 JLF15 0.41±0.10a 0.52±0.10ab 0.55±0.16ab 1.11±0.27b 1.16±0.29b 0.98±0.26ab 0.59 Values with a different letter superscript are significantly different from each other (P ≤ 0.05): x, y indicate difference between farrowing systems on the same sampling day; a, b indicate difference between sampling days in the same farrowing system. a -1D: 1 day before crating. b + 1D: 1 day after crating. c Day 2: 1 day before crate opening. d Day 4: 1 day after crate opening. Table 5. CORT and CgA in FC remained similar throughout the lactation further insight on the feasibility of TC on commercial conditions in period. However, CORT in SWAP peaked on D2, compared to + 1D warm climates. crating (P = 0.02). Additionally, CgA in JLF15 peaked on D4 and In terms of the difference between two TC, the creep area of SWAP is remained elevated on D12, compared to − 1D crating (P = 0.05 and designed near the head of the sow to facilitate a ‘nest-like’ situation. 0.02, respectively). SWAP pen is equipped with a sloping wall, as it is preferred by the sows when they lie down (Damm et al., 2006), to create an environment that 3.3. Number of crushed piglets per sow and parameters for analyzing the is preferred by the sow and protects the piglets. In addition, the design of crushing events SWAP is based on a loose housed system where a simple type of confinement – only one wing – is implemented. On the other hand, Number of crushed piglets per sow between farrowing systems were JLF15 is based on the traditional crate system where there are two wings as follows: 0.3 ± 0.1 in FC, 1.2 ± 0.3 in SWAP, and 0.6 ± 0.2 in JLF15. but allowing it to open up for the sows to move around. In JLF15, there is Crushing rate of SWAP was higher than FC (P < 0.0001) and JLF15 no preferred support for the sows but the wings and the rail are the (P = 0.02), and that of JLF15 was also higher than FC (P < 0.0001). design for piglet protection for the sows to lean against when they lie There were in total 49 crushed piglets during the study period. down. The design of JLF15 is not based on the sows’ biological needs Table 6 lists the details of the parameters used for analyzing these 49 and has no preferred lying or dunging areas like SWAP where they can crushing events. Crushing rate in the autumn batch was significantly divide in zones. lower than in other batches, including the winter batch (P < 0.0001), the spring batch (P < 0.0001), and the summer batch (P = 0.007). As 4.1. Mother-young interactions, social interactions between piglets, and shown in Table 6, crushing events occurred similarly before and after exploration in piglets and sows opening the crate in SWAP and JLF15 (P = 0.54). Additionally, almost two third (63.3%) of the crushing events occurred when the sows used On D4, one day after the sows were set loose from the temporary an aid from the pen while changing posture, and the percentage in FC crates, SWAP and JLF15 sows were exploring the pens 10-times and six- was higher than in SWAP and in JLF15 (P = 0.05). times more than FC sows respectively. Exploration in SWAP and JLF15 sows was similar due to similar setups in TC (i.e. similar space allowance 3.4. Growth performance and prevalence of foreleg abrasion in piglets and access to the nest-building materials). Exploration in TC sows also reached the peak on D4. This finding is in agreement with Chidgey et al. Growth performance in piglets is presented in Table 7. No difference (2016) and Goumon et al. (2018), where exploration is highly motivated was found between farrowing systems in BW3 (P = 0.71 and 0.48), BW19 due to curiosity when the environment changes (e.g. increased space (P = 0.28 and 0.41), and ADG3–19 (P = 0.23 and 0.43), using either allowance, presence of newborn piglets) (Wood-Gush and Vestergaard, piglet or litter as the experimental unit respectively. BW3 had an effect 1989). In mid-lactation, SWAP and JLF15 sows interacted with the on BW19 and ADG3–19 (both P < 0.05). piglets six-times and five-times more than FC sows respectively, which Prevalence of foreleg abrasion on D19 in piglets between farrowing also agrees with the studies of Chidgey et al. (2016) and Singh et al. systems was as follows: 66.3 ± 3.6% (n = 175) in FC, 67.5 ± 3.2% in (2017). As Chidgey et al. (2017) stated, mother-young interactions are SWAP (n = 209), and 73.5 ± 3.0% in JLF15 (n = 219). No difference determined by the farrowing environment. Due to a larger and open was found between farrowing systems in the prevalence of foreleg floor space without the restriction of the crate structure in both TC, loose abrasion (P = 0.28). sows can better orient themselves towards piglets. Sow-piglet nose contacts occur frequently when they are in an open space (Portele et al., 4. Discussion 2019), and the result of NNC (i.e. piglet-initiated naso-naso contact with the sow) confirms the findings by Jarvis et al. (2004) and Bolhuis et al. In the present study, we compared the behavior, stress physiology (2018), in which they found that piglets housed in loose house systems and performance of sows and piglets in three farrowing systems, had more NNC with their mother sows. Although all the sows were including one conventional farrowing crate system (FC), and two far­ confined on D2, a narrower crate length and width in FC may force the rowing pens with temporary crating system (TC, i.e. SWAP and JLF15) sows to lay under the feed trough which makes it difficult for FC piglets which are available in the market. The study was conducted on a com­ to express NNC. mercial farm in Spain, in which the results are expected to provide On the other hand, FC piglets interacted with littermates more than 6 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 Table 6 Table 7 Parameters used for analyzing all the crushing events (n = 49) in the three Body weight (kg) on Day 3 (BW3) and 19 (BW19), and average daily gain farrowing systems during the study period (from November 2018 to July 2019) (ADG3–19) (g/day) of piglets in the three farrowing systems during the lactation (FC: conventional farrowing crate; SWAP and JLF15: farrowing pens with period (FC: conventional farrowing crate; SWAP and JLF15: farrowing pens with temporary crating). These crushing events all led to the death of the piglets, temporary crating). which was confirmed by the surveillance cameras. Except crushing rate, data are Farrowing system n Mean SEM P-value presented in number of piglets. The sum of each parameter in each farrowing system adds up to the number of crushed piglets in each farrowing system. Using piglet as the experimental unit BW3, kg FC 183 2.02 0.03 0.71 FC SWAP JLF15 SWAP 243 1.88 0.03 Number of crushed piglets 6 28 15 JLF15 237 1.97 0.03 Number of live born piglets 191 252 259 BW19, kg FC 175 5.11 0.08 0.28 Crushing rate (%) in each system 3.1 11.1 5.8 SWAP 209 5.21 0.09 Parameters FC SWAP JLF15 P-value JLF15 219 5.15 0.08 Batch1 0 3 4 < 0.0001 ADG3–19, g/day FC 175 176.88 4.20 0.23 Autumn 4 7 4 SWAP 206 189.49 4.77 Winter 1 6 4 JLF15 219 179.99 4.38 Spring 1 12 3 Using litter as the experimental unit Summer BW3, kg FC 18 2.02 0.06 0.48 Parity1 1 3 1 0.74 SWAP 23 1.89 0.05 Primiparous 5 25 14 JLF15 23 1.98 0.07 Multiparous BW19, kg FC 18 5.12 0.15 0.41 Crushing time2 2 9 8 0.30 SWAP 23 5.20 0.20 Daytime (07:01–18:00 h) 3 19 6 JLF15 23 5.16 0.14 Night (18:01–07:00 h) 1 0 1 ADG3–19, g/day FC 18 177.49 8.55 0.43 Unknown SWAP 23 186.57 10.80 Sow was crated/loosea,3 6 9 7 0.54 JLF15 23 180.25 9.26 Crated - 19 8 Loose Crushing dayb,2 4 10 7 0.52 which may also suggest an inadequate FC pen size for the piglets. As Before Day 3 0 5 1 Day 3 2 13 7 reported by Turner et al. (2000), number of skin lesions increased in pigs After Day 3 housed in low space allowance, indicating an increased aggression Body part of the sow2 1 1 1 0.49 within the group; aggression is considered as a SB in the present study. Front (head to front legs) 2 16 10 Regarding PPE (i.e. piglet locomotor or object play, and exploration Middle (trunk) 2 11 3 of the pen), we did not observe any difference between farrowing sys­ Back (hind legs to tail) 1 0 1 Unknown tems in the present study. However, Singh et al. (2017) found an Posturec change – from posture 12 4 13 5 0.33 increased play behavior and a reduced manipulative behavior (i.e. Stand 0 6 1 manipulative behavior would be considered as a SB in our study, a Sternal/ventral recumbency 1 3 4 negative social interaction) in loose pen piglets. Chaloupková et al. Sit 0 6 4 Lateral recumbency (i.e. lie) 1 0 1 (2007) and Oostindjer et al. (2011) also found an increased play Unknown behavior in piglets living in an enriched and enlarged pre-weaning Posture change – to posture 22 0 0 0 0.46 environment. However, in our study, locomotor play, object play, and Stand 4 14 8 exploration were not distinguished, so it would be difficult to draw any Sternal/ventral recumbency 0 0 0 conclusion on the effect of farrowing systems on play behavior. In Sit 1 14 6 Lateral recumbency (i.e. lie) 1 0 1 addition, apart from an enlarged space, SWAP and JLF15 piglets did not Unknown receive regular enrichment (i.e. straw bedding) like in Chaloupková Did the sow use the pen fixture as an aid 4 8 3 0.05 et al. (2007). Oostindjer et al. (2011) also stated that development of to change her posture?d,2 1 20 10 foraging related behaviors was positively influenced by early environ­ Yes 1 0 2 mental enrichment, but less by sow housing. No Unknown a FC sows could only be crated during the study. 4.2. Salivary stress biomarkers in sows during lactation b Day 0: the day of farrowing. Day 3: Opening of the crates in SWAP and JLF15 pens. In the present study, we did not find any difference in salivary stress c List of sow postures is adopted from Wischner et al. (2009). d biomarkers, cortisol (CORT) and chromogranin A (CgA), between far­ FC sows could only use the crate structure to change the posture during the rowing systems during lactation, except CORT in SWAP on D2, which study. 1 Analyzed by a general linear model with a Poisson distribution: number of was higher than in FC and JLF15 on the same day. This lack of significant crushed piglets per sow as the response variable, farrowing system, batch and differences between farrowing systems suggests that both temporary parity time as the fixed effects, and litter size as the covariate. confinement (no difference between − 1D and +1D crating) and removal 2,3 Analyzed with chi-square tests: comparing the percentage of occurrence of of confinement (no difference between D2 and D4) did not elevate the each parameter between 2FC, SWAP and JLF15; between 3SWAP and JLF15. stress level (i.e. similar adrenal reactivity) in sows in SWAP and JLF15. Similarly, Goumon et al. (2018) did not find the effect of removal of confinement after 24 h in CORT. Level of CORT seems inconclusive in sows in early lactation as some studies found no difference between farrowing crate and farrowing pens (Cronin et al., 1992; Biensen et al., TC piglets on D2 and D4. As we found negative correlations between SB 1996), whereas other studies found a higher (Oliviero et al., 2008) or (i.e. social interactions between piglets) vs. NNC and MYI (i.e. sow lower (Hales et al., 2016) level of CORT in crated sows. As CORT is initiated physical contact with the piglet), higher SB in FC might suggest known to be released quickly within 30 min after a stressor is intro­ a different time budget in those piglets as they spent more time inter­ duced, the fact that we did not see a change in CORT after temporary acting with their littermates instead of the sow, due to the structure of confinement or removal of confinement could be due to a short-lasting the crate. Moreover, SB may increase when the space allowance is small, effect (Goumon et al., 2018). A higher CORT in SWAP sows on D2 7 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 may be linked to the sow’s inability to avoid piglets’ call for nursing (a urinating, and rest outside with the sow, and thus increased the risk of higher frequency of NNC in SWAP on D2) due to crating, as shown in the being crushed (Marchant et al., 2001). It is possible that there were more crated sows in Oliviero et al. (2008). It may reflect that nursing is litters in SWAP resting outside than JLF15 litters, as the percentage of difficult for both sows and piglets when the sow is lying with the udder crushing in the summer batch in SWAP is twice more than in JLF15, and facing towards the sloping wall, making the space for nursing/suckling thus resulting in a higher percentage in SWAP than in JLF15 in average crowded. even after the crate opening. Although we did not find any difference in A peak of CgA in JLF15 sows on D4 and D12 suggested that there was RSC between farrowing systems on the same observation day, it could be a stress response which activated the SAM (sympathetic-adreno- that we only counted the piglets that were in actual physical contact medullar) axis (Ott et al., 2014) before saliva sampling. Few studies with the sow. Therefore, unless crowding in piglets occurred and most of investigated the level of change in CgA after an acute stressor was the piglets were in physical contact with the sow, the numbers we applied in pigs (e.g. immobilization/nose snare, Escribano et al., 2013 collected could not reflect the crowding situation well. It is thus indis­ and Huang et al., 2017; mixing and feed deprivation, Ott et al., 2014; pensable to adjust the temperature of the creep area regularly to psychosocial stress (i.e. regrouping or isolation), Escribano et al., 2015; encourage piglets resting inside to avoid being crushed. weaning, Ko et al., 2020), but these acute stressors did not apply to our study. Additionally, CORT in JLF15 sows also showed a peak on D4 and 4.4. Growth performance and foreleg abrasion in piglets D12 (but not significant), which might indicate that some uncontrollable stressor occurred before saliva sampling that cannot be confirmed from Unlike previous studies (Pedersen et al., 2011; Melišová et al., 2014), the rest of the results. This unknown stressor (e.g. from handling or from we did not find any effect of the farrowing system on weight gain in the environment) might have had activated the SAM axis (CgA as the piglets, even though there was a 211 g (from D3 to D19) of difference indicator) to a greater extent than the HPA (hypothal­ between FC and SWAP numerically. Similar growth performance may amic–pituitary–adrenal) axis (CORT as the indicator), which altered the suggest an undisturbed suckling and nursing behaviors in a short- and CgA level significantly more than the CORT level. Further research long-term with regard to crate removal (Goumon et al., 2018), but the which also analyzes different salivary biomarkers, such as IgA and amount of creep feed intake may also contribute to pre-weaning weight oxytocin, is recommended to better understand the general reactivity of gain (Oostindjer et al., 2010), which is a parameter we did not measure the stress level in sows in FC and TC during lactation. in the present study. Prevalence of foreleg abrasion in piglets in the present study was 4.3. Crushing in farrowing crates vs. farrowing pens with temporary between 66% and 73%, which was relatively high, compared to Mout­ crating system totou et al. (1999) (36%, highly associated with part-concrete, par­ t-round-mesh flooring) in the UK, Hoy et al. (1999) (54–84%) in Sows in TC were set loose from D3 (i.e. the 4th day postpartum), Germany, and Johansen et al. (2004) (46%, cast iron slats) in Denmark. which was a day earlier than Moustsen et al. (2013) recommended, but Cushioning and minimizing friction with mineral oil as Goumon et al. (2018) suggested. In the present study, both TC showed impregnated-neoprene sponge at the suckling flooring area were re­ a higher crushing rate than FC during lactation. A higher number of ported to greatly reduce leg injuries (Phillips et al., 1995). More research crushed piglets in SWAP is likely to reflect the fact that the crate was focusing on optimal flooring design for indoor farrowing systems to wider than both FC and JLF15, meaning that SWAP sows were not as avoid the development of foreleg abrasion is needed. restricted as FC and JLF15 sows. A sloping wall on one side in SWAP could also block the piglets to escape from the sows if the piglets were 5. Conclusions resting at the wall side. On the other hand, in FC and JLF15, there was relatively more space on both sides of the sows. Crushing rate in TC was Farrowing pens with temporary crating system enhanced sow-piglet similar before and after D3, which was different from Chidgey et al. interactions and sow explorative behavior. Neither the farrowing system (2015) and King et al. (2019), where both studies reported an increased per se nor the opening of the temporary crates altered sows’ salivary crushing rate after crate opening. One of the reasons for this could be stress biomarkers. Crushing rate was higher in temporary crated sows due to a relatively low number of animals involved in the study (64 sows than in crates in the summer batch, emphasizing the importance of a in the present study; 732 sows in Chidgey et al., 2015; 168 sows in King correct management of the piglets’ creep area. Average daily gain in et al., 2019). Additionally, different from most of the studies which used piglets during lactation was similar across farrowing systems. Farrowing hyper-prolific sows with lower birth weight piglets, our study used a pens with temporary crating system are feasible to house crossbred rustic breed with an average of 10–11 piglets born alive with a relatively Duroc sows and piglets on commercial farms in the Mediterranean re­ high birth weight. As Melišová et al. (2011) stated, piglets in better body gion. Knowledge-exchange strategies on sow and piglets management condition can afford to stay close the sow without being crushed, which around farrowing should be further enhanced to ensure the suitability of may be the case in our study. temporary crating systems under commercial conditions in Spain and In the present study, piglet crushing mostly occurred in FC in the elsewhere. winter batch and in SWAP in the summer batch, even though the per­ centages were very low. However, most of the crushing events in FC Funding happened even when the sows used the crate as a support to change their posture. Heat sources for piglets in the farrowing systems generally The study was funded by Proyectos I+D+I Convocatoria RETOS include the heating mat, the lamp and sow’s udder. Lack of space in FC 2015 (AGL2015-68373-C2-1-R) from the Spanish Ministry of Economy, may result in crowding (Rangstrup-Christensen et al., 2018) in winter, Industry and Competitiveness (MINECO), and the Technological where piglets occupied the heating mat under the lamp and rested next Transfer of the Rural Development Program 2014–2020 from the to sow’s udder, making the sow difficult to change her posture without Catalan Government. HLK received the University, Research Center and overlaying the piglets, even using the pen fixture (i.e. crate in this case) Hospital Foundation Grants for the Contracting of New Research Staff as a support. On the contrary, piglets in SWAP and JLF15 were able to (FI) from Agency for Management of University and Research Grants rest in the creep area with the heating mats and the lamp in winter (AGAUR) of the Catalan Government. PL received the Juan de la Cierva where the sow had no access to, possibly reducing the chance of being postdoctoral fellowship from the Spanish Ministry for Science and overlaid by the sow. On the other hand, due to incorrect management, Innovation (MICINN). the temperature in the creep area in summer might have been too high for piglets, which made them use the creep area for defecating and 8 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 Conflicts of interest Hales, J., Moustsen, V.A., Nielsen, M.B.F., Hansen, C.F., 2014. Higher preweaning mortality in free farrowing pens compared with farrowing crates in three commercial pig farms. Animal 8, 113–120. https://doi.org/10.1017/ Janni Hales works at SKIOLD JYDEN which produces the SWAP and S1751731113001869. JLF15 pens, but only participated in the housing design of the SWAP and Hales, J., Moustsen, V.A., Nielsen, M.B.F., Hansen, C.F., 2015. Temporary confinement of JLF15 pens of the study, interpretation of data, and assisted in the loose-housed hyperprolific sows reduces piglet mortality. J. Anim. Sci. 93, 4079–4088. https://doi.org/10.2527/jas.2015-8973. writing of the manuscript. The funders had no role in the design of the Hales, J., Moustsen, V.A., Nielsen, M.B.F., Hansen, C.F., 2016. The effect of temporary study; in the collection, analyses, or interpretation of data; in the writing confinement of hyperprolific sows in Sow Welfare and Piglet protection pens on sow of the manuscript, or in the decision to publish the results. behaviour and salivary cortisol concentrations. Appl. Anim. Behav. Sci. 183, 19–27. https://doi.org/10.1016/j.applanim.2016.07.008. Hansen, C.F., Hales, J., Weber, P.M., Edwards, S.A., Moustsen, V.A., 2017. Confinement Acknowledgements of sows 24h before expected farrowing affects the performance of nest building behaviours but not progress of parturition. Appl. Anim. Behav. Sci. 188, 1–8. https:// doi.org/10.1016/j.applanim.2017.01.003. The authors gratefully acknowledge all the collaborators who made Hoy, S., Ziron, M., Iben, B., 1999. Origin and effects of skin erosion in piglets and this study possible, including Gerfam Rubirola, L′ Ekip, Jyden and possibilities for their prevention. Prakt. Tierärzt. 80, 698–706. Rotecna. We thank Dr Eva Mainau, Dr Josep Gasa, and Dr David Solà for Huang, Y., Liu, Z., Liu, W., Yin, C., Ci, L., Zhao, R., Yang, X., 2017. Salivary haptoglobin and chromogranin A as non-invasive markers during restraint stress in pigs. Res. Vet. providing advice on experimental design, Francisco Benitez for partici­ Sci. 114, 27–30. https://doi.org/10.1016/j.rvsc.2017.02.023. pating in the pilot study, David Arias, Gabriela Rojas, Alejandra Feld, Jarvis, S., Reed, B.T., Lawrence, A.B., Calvert, S.K., Stevenson, J., 2004. Peri-natal and Lourdes Llonch for assisting with data collection during the study environmental effects on maternal behaviour, pituitary and adrenal activation, and the progress of parturition in the primiparous sow. Anim. Welf. 13 (2), 171–181. period. We also thank the statistical consultation provided by Applied Jensen, P., 1986. Observations on the maternal behaviour of free-ranging domestic pigs. Statistics Service (SEA) of UAB, and the technical support from Appl. Anim. Behav. Sci. 16 (2), 131–142. https://doi.org/10.1016/0168-1591(86) Department of Animal and Food Science of UAB and Interdisciplinary 90105-X. Johansen, M., Alban, L., Kjærsgård, H.D., Bækbo, P., 2004. Factors associated with Laboratory of Clinical Analysis of University of Murcia. suckling piglet average daily gain. Prev. Vet. Med. 63 (1–2), 91–102. https://doi. org/10.1016/j.prevetmed.2004.01.011. References Johnson, A.K., Marchant-Forde, J.N., 2009. Welfare of pigs in the farrowing environment. In: The Welfare of Pigs. Springer, Dordrecht, the Netherlands, pp. 141–188. https://doi.org/10.1007/978-1-4020-8909-1_5. Andersen, I.L., Berg, S., Bøe, K.E., 2005. Crushing of piglets by the mother sow (Sus King, R.L., Baxter, E.M., Matheson, S.M., Edwards, S.A., 2019. Temporary crate opening scrofa)—purely accidental or a poor mother? Appl. Anim. Behav. Sci. 93 (3–4), procedure affects immediate post-opening piglet mortality and sow behaviour. 229–243. https://doi.org/10.1016/j.applanim.2004.11.015. Animal 13 (1), 189–197. https://doi.org/10.1017/S1751731118000915. Baxter, E.M., Lawrence, A.B., Edwards, S.A., 2012. Alternative farrowing Ko, H.L., Chong, Q., Escribano, D., Camerlink, I., Manteca, X., Llonch, P., 2020. Pre- accommodation: welfare and economic aspects of existing farrowing and lactation weaning socialization and environmental enrichment affect life-long response to systems for pigs. Animal 6 (1), 96–117. https://doi.org/10.1017/ regrouping in commercially-reared pigs. Appl. Anim. Behav. Sci. 229, 105044 S1751731111001224. https://doi.org/10.1016/j.applanim.2020.105044. Baxter, E.M., Andersen, I.L., Edwards, S.A., 2018. Sow welfare in the farrowing crate and Lohmeier, R.Y., Grimberg-Henrici, C.G.E., Büttner, K., Burfeind, O., Krieter, J., 2020. alternatives. In: Advances in Pig Welfare. Woodhead Publishing, Duxford, United Farrowing pens used with and without short-term fixation impact on reproductive Kingdom, pp. 27–72. https://doi.org/10.1016/B978-0-08-101012-9.00002-2. traits of sows. Livest. Sci. 231, 103889 https://doi.org/10.1016/j. Biensen, N.J., von Borell, E.H., Ford, S.P., 1996. Effects of space allocation and livsci.2019.103889. temperature on periparturient maternal behaviors, steroid concentrations, and piglet Marchant, J.N., Broom, D.M., Corning, S., 2001. The influence of sow behaviour on piglet growth rates. J. Anim. Sci. 74 (11), 2641–2648. https://doi.org/10.2527/ mortality due to crushing in an open farrowing system. J. Anim. Sci. 72 (1), 19–28. 1996.74112641x. https://doi.org/10.1017/S135772980005551X. Bolhuis, J.E., Raats-van den Boogaard, A.M.E., Hoofs, A.I.J., Soede, N.M., 2018. Effects of Martin, J.E., Ison, S.H., Baxter, E.M., 2015. The influence of neonatal environment on loose housing and the provision of alternative nesting material on peri-partum sow piglet play behaviour and post-weaning social and cognitive development. Appl. behaviour and piglet survival. Appl. Anim. Behav. Sci. 202, 28–33. https://doi.org/ Anim. Behav. Sci. 163, 69–79. https://doi.org/10.1016/j.applanim.2014.11.022. 10.1016/j.applanim.2018.01.004. Melišová, M., Illmann, G., Andersen, I.L., Vasdal, G., Haman, J., 2011. Can sow pre-lying Chaloupková, H., Illmann, G., Bartoš, L., Špinka, M., 2007. The effect of pre-weaning communication or good piglet condition prevent piglets from getting crushed? Appl. housing on the play and agonistic behaviour of domestic pigs. Appl. Anim. Behav. Anim. Behav. Sci. 134 (3–4), 121–129. https://doi.org/10.1016/j. Sci. 103 (1–2), 25–34. https://doi.org/10.1016/j.applanim.2006.04.020. applanim.2011.06.015. Chidgey, K.L., Morel, P.C., Stafford, K.J., Barugh, I.W., 2015. Sow and piglet productivity Melišová, M., Illmann, G., Chaloupková, H., Bozděchová, B., 2014. Sow postural and sow reproductive performance in farrowing pens with temporary crating or changes, responsiveness to piglet screams, and their impact on piglet mortality in farrowing crates on a commercial New Zealand pig farm. Livest. Sci. 173, 87–94. pens and crates. J. Anim. Sci. 92 (7), 3064–3072. https://doi.org/10.2527/jas.2013- https://doi.org/10.1016/j.livsci.2015.01.003. 7340. Chidgey, K.L., Morel, P.C., Stafford, K.J., Barugh, I.W., 2016. Observations of sows and Mellor, D.J., Patterson-Kane, E., Stafford, K.J., 2009. In: Agricultural Sciences and piglets housed in farrowing pens with temporary crating or farrowing crates on a Animal Welfare, first ed. The Sciences of Animal Welfare, Herts, UK, pp. 15–33. commercial farm. Appl. Anim. Behav. Sci. 176, 12–18. https://doi.org/10.1016/j. Moustsen, V.A., Hales, J., Lahrmann, H.P., Weber, P.M., Hansen, C.F., 2013. applanim.2016.01.004. Confinement of lactating sows in crates for 4 days after farrowing reduces piglet Chidgey, K.L., Morel, P.C.H., Stafford, K.J., Barugh, I.W., 2017. Sow and piglet mortality. Animal 7 (4), 648–654. https://doi.org/10.1017/S1751731112002170. behavioral associations in farrowing pens with temporary crating and in farrowing Mouttotou, N., Hatchell, F.M., Green, L.E., 1999. The prevalence and risk factors crates. J. Vet. Behav. 20, 91–101. https://doi.org/10.1016/j.jveb.2017.01.003. associated with forelimb skin abrasions and sole bruising in preweaning piglets. Cronin, G.M., Smith, J.A., 1992. Suckling behaviour of sows in farrowing crates and Prev. Vet. Med. 39 (4), 231–245. https://doi.org/10.1016/S0167-5877(99)00006-9. straw-bedded pens. Appl. Anim. Behav. Sci. 33 (2–3), 175–189. https://doi.org/ Nicolaisen, T., Lühken, E., Volkmann, N., Rohn, K., Kemper, N., Fels, M., 2019. The effect 10.1016/S0168-1591(05)80006-1. of sows’ and piglets’ behaviour on piglet crushing patterns in two different farrowing Damm, B.I., Moustsen, V., Jørgensen, E., Pedersen, L.J., Heiskanen, T., Forkman, B., pen systems. Animals 9 (8), 538. https://doi.org/10.3390/ani9080538. 2006. Sow preferences for walls to lean against when lying down. Appl. Anim. Nowland, T.L., van Wettere, W.H.E.J., Plush, K.J., 2019. Allowing sows to farrow Behav. Sci. 99 (1–2), 53–63. https://doi.org/10.1016/j.applanim.2005.09.014. unconfined has positive implications for sow and piglet welfare. Appl. Anim. Behav. Escribano, D., Fuentes-Rubio, M., Cerón, J.J., 2012. Validation of an automated Sci. 221, 104872 https://doi.org/10.1016/j.applanim.2019.104872. chemiluminescent immunoassay for salivary cortisol measurements in pigs. J. Vet. Oliviero, C., Heinonen, M., Valros, A., Hälli, O., Peltoniemi, O.A.T., 2008. Effect of the Diagn. Invest. 24, 918–923. https://doi.org/10.1177/1040638712455171. environment on the physiology of the sow during late pregnancy, farrowing and Escribano, D., Soler, L., Gutiérrez, A.M., Martínez-Subiela, S., Cerón, J.J., 2013. early lactation. Anim. Reprod. Sci. 105 (3–4), 365–377. https://doi.org/10.1016/j. Measurement of chromogranin A in porcine saliva: validation of a time-resolved anireprosci.2007.03.015. immunofluorometric assay and evaluation of its application as a marker of acute Oliviero, C., Heinonen, M., Valros, A., Peltoniemi, O., 2010. Environmental and sow- stress. Animal 7, 640–647. https://doi.org/10.1017/S1751731112002005. related factors affecting the duration of farrowing. Anim. Reprod. Sci. 119 (1–2), Escribano, D., Gutiérrez, A.M., Tecles, F., Cerón, J.J., 2015. Changes in saliva biomarkers 85–91. https://doi.org/10.1016/j.anireprosci.2009.12.009. of stress and immunity in domestic pigs exposed to a psychosocial stressor. Res. Vet. Oostindjer, M., Bolhuis, J.E., Mendl, M., Held, S., Gerrits, W., van den Brand, H., Sci. 102, 38–44. https://doi.org/10.1016/j.rvsc.2015.07.013. Kemp, B., 2010. Effects of environmental enrichment and loose housing of lactating Goumon, S., Leszkowová, I., Šimečková, M., Illmann, G., 2018. Sow stress levels and sows on piglet performance before and after weaning. J. Anim. Sci. 88 (11), behavior and piglet performances in farrowing crates and farrowing pens with 3554–3562. https://doi.org/10.2527/jas.2010-2940. temporary crating. J. Anim. Sci. 96 (11), 4571–4578. https://doi.org/10.1093/jas/ Oostindjer, M., van den Brand, H., Kemp, B., Bolhuis, J.E., 2011. Effects of environmental sky324. enrichment and loose housing of lactating sows on piglet behaviour before and after Gu, Z., Gao, Y., Lin, B., Zhong, Z., Liu, Z., Wang, C., Li, B., 2011. Impacts of a freedom weaning. Appl. Anim. Behav. Sci. 134 (1–2), 31–41. https://doi.org/10.1016/j. farrowing pen design on sow behaviours and performance. Prev. Vet. Med. 102 (4), applanim.2011.06.011. 296–303. https://doi.org/10.1016/j.prevetmed.2011.08.001. 9 H.-L. Ko et al. Applied Animal Behaviour Science 246 (2022) 105527 Ott, S., Soler, L., Moons, C.P.H., Kashiha, M.A., Bahr, C., Vandermeulen, J., Janssens, S., Singh, C., Verdon, M., Cronin, G.M., Hemsworth, P.H., 2017. The behaviour and welfare Gutiérrez, A.M., Escribano, D., Cerón, J.J., Berckmans, D., Tuyttens, F.A.M., of sows and piglets in farrowing crates or lactation pens. Animal 11 (7), 1210–1221. Niewold, T.A., 2014. Different stressors elicit different responses in the salivary https://doi.org/10.1017/S1751731116002573. biomarkers cortisol, haptoglobin, and chromogranin A in pigs. Res. Vet. Sci. 97 (1), Turner, S.P., Ewen, M., Rooke, J.A., Edwards, S.A., 2000. The effect of space allowance 124–128. https://doi.org/10.1016/j.rvsc.2014.06.002. on performance, aggression and immune competence of growing pigs housed on Pedersen, L.J., Damm, B.I., Marchant-Forde, J.N., Jensen, K.H., 2003. Effects of feed- straw deep-litter at different group sizes. Livest. Prod. Sci. 66 (1), 47–55. https://doi. back from the nest on maternal responsiveness and postural changes in primiparous org/10.1016/S0301-6226(00)00159-7. sows during the first 24h after farrowing onset. Appl. Anim. Behav. Sci. 83 (2), Wischner, D., Kemper, N., Stamer, E., Hellbruegge, B., Presuhn, U., Krieter, J., 2009. 109–124. https://doi.org/10.1016/S0168-1591(03)00116-3. Characterisation of sows’ postures and posture changes with regard to crushing Pedersen, M.L., Moustsen, V.A., Nielsen, M.B.F., Kristensen, A.R., 2011. Improved udder piglets. Appl. Anim. Behav. Sci. 119 (1–2), 49–55. https://doi.org/10.1016/j. access prolongs duration of milk letdown and increases piglet weight gain. Livest. applanim.2009.03.002. Sci. 140 (1–3), 253–261. https://doi.org/10.1016/j.livsci.2011.04.001. Wood-Gush, D.G.M., Vestergaard, K., 1989. Exploratory behavior and the welfare of Phillips, P.A., Fraser, D., Pawluczuk, B., 1995. Effects of cushioned flooring on piglet leg intensively kept animals. J. Agric. Environ. Ethics 2, 161–169. https://doi.org/ injuries. Trans. ASAE 38 (1995), 213–216. https://doi.org/10.13031/2013.27832. 10.1007/BF01826929. Portele, K., Scheck, K., Siegmann, S., Feitsch, R., Maschat, K., Rault, J.L., Camerlink, I., Yun, J., Swan, K.-M., Vienola, K., Farmer, C., Oliviero, C., Peltoniemi, O., Valros, A., 2019. Sow-piglet nose contacts in free-farrowing pens. Animals 9 (8), 513. https:// 2013. Nest-building in sows: effects of farrowing housing on hormonal modulation doi.org/10.3390/ani9080513. of maternal characteristics. Appl. Anim. Behav. Sci. 148 (1–2), 77–84. https://doi. Rangstrup-Christensen, L., Krogh, M.A., Pedersen, L.J., Sørensen, J.T., 2018. Sow level org/10.1016/j.applanim.2013.07.010. risk factors for early piglet mortality and crushing in organic outdoor production. Animal 12 (4), 810–818. https://doi.org/10.1017/S1751731117002178. 10