Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing

G Model

ARTICLE IN PRESS

PREVET-3761; No. of Pages 12

Preventive Veterinary Medicine xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed

Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing Rebecka Westin a,b,∗ , Nils Holmgren b , Jan Hultgren a , Kerstin Ortman c , Anders Linder c , Bo Algers a a b c

Department of Animal Environment and Health, Swedish University of Agricultural Sciences, P.O. Box 234, SE-532 23 Skara, Sweden Swedish Animal Health Service, SE-532 89 Skara, Sweden Eurofins Food & Agro Testing Sweden AB, P.O. Box 399, SE-532 24 Skara, Sweden

a r t i c l e

i n f o

Article history: Received 11 September 2014 Received in revised form 9 February 2015 Accepted 21 February 2015 Keywords: Piglet survival Stillborn Starvation Crushing Hypothermia New neonatal porcine diarrhoea

a b s t r a c t Piglet survival is the outcome of complex interactions between the sow, the piglet and their environment. In order to facilitate nest-building and to provide a suitable environment for the newborn piglets, a strategic method to supply loose housed sows with large quantities of straw at farrowing has been developed by Swedish piglet-producing farmers. The objectives of this cohort study were to use post-mortem findings to assess the causes of death and to quantify the effect of a large quantity of straw provided before farrowing compared to limited small daily amounts on stillbirths, post-mortem findings in piglets dying within 5 days after birth and the pre-weaning mortality. On each of four commercial piglet-producing farms in South-West Sweden, one batch of sows was studied during two consecutive lactations. At inclusion, sows were randomly assigned to two treatment groups, and sows remaining in the batch during the next lactation switched treatment group. In the STRAW group (n = 181 litters) sows were provided with 15–20 kg of chopped straw 2 days prior to the calculated date of farrowing. Sows in the CONTROL group (n = 182 litters) received 0.5–1 kg of chopped straw on a daily basis plus about 2 kg for nest-building when the stockperson judged the sow to be about to farrow. After onset of farrowing, additionally 1–2 kg was given. Post-mortem examination was performed in all piglets that died within 5 days after birth (n = 798). The three major post-mortem findings were starvation (34%) crushing by the sow (28%), and enteritis (24%). In conclusion, strategic use of large quantities of straw reduced the number of stillborn piglets per litter by 27% (p = 0.007). Under the conditions studied, the pre-weaning mortality of liveborn piglets was not affected by treatment; however, the distribution of post-mortem findings differed with fewer piglets dying due to starvation and more due to crushing and enteritis in STRAW litters. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Piglet survival is the outcome of complex interactions between the sow, the piglet and the environment

∗ Corresponding author at: Department of Animal Environment and Health, Swedish University of Agricultural Sciences, P.O. Box 234, SE-532 23 Skara, Sweden. Tel.: +46 51167218; Mobile phone: +46 5111067204. E-mail address: [email protected] (R. Westin).

(Edwards, 2002), and high pre-weaning mortality is a wellknown phenomenon within piglet production. Starvation and crushing, both closely linked to perinatal asphyxia and hypothermia, are the most common ultimate causes of death in liveborn piglets (reviewed by: Alonso-Spilsbury et al., 2007; Edwards, 2002; Herpin et al., 2002). Through the decades, different approaches have been tried in order to reduce piglet mortality. Commercial strategies have often focused on modifying the farrowing

http://dx.doi.org/10.1016/j.prevetmed.2015.02.023 0167-5877/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

2

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

environment to affect sow behaviour and increase human intervention (Edwards, 2002). In spite of this, higher numbers of piglets weaned are still being achieved by increasing prolificacy rather than by reducing mortality (Kirkden et al., 2013). The amount of research focusing on individual piglet traits and their importance for survival is increasing. Studies focusing on maternal abilities and management routines, such as provision of nest material, are less frequent. An association between increased nest-building and reduced mortality has been found in some studies (Andersen et al., 2005; Cronin and van Amerongen, 1991) but not in others (Cronin and Smith, 1992; Edwards and Furniss, 1988). In Sweden, piglet-producing farmers have developed a method for a strategic use of large quantities of straw at farrowing for loose housed sows. These farmers supply their sows with 15–20 kg of chopped straw once at 2 days prior to the calculated date of farrowing. Gradually the straw drains through the slatted floor (Westin et al., 2013) and is then replaced by a daily supply of 0.5–1 kg straw in accordance with common Swedish management routines. One purpose is to facilitate nest-building for the sow. A second goal is to provide a suitable environment for the piglets at birth. The method is proven to be technically feasible (Westin et al., 2013), to prevent bruising on claws and limbs and increase weight gain in piglets (Westin et al., 2014) and to increase the amount of nest-building performed pre-partum (Westin et al., 2015). Oxytocin plays an important part in termination of the nest building phase as well as in the farrowing process and lactation in sows (Algers and Uvnäs-Moberg, 2007). Oxytocin also stimulates maternal interaction and attachment between mother and young (Uvnäs-Moberg et al., 2001). Use of farrowing crates during nest building and farrowing has been demonstrated to cause physiological stress responses in sows (Jarvis et al., 1997; Oliviero et al., 2008) and to prolong the farrowing duration possibly through an inhibition of oxytocin (Oliviero et al., 2008). In the present study we hypothesised that permitting the sow to express her natural nest-building behaviour by introduction of “strategic use of straw” will reduce the sow’s potential stress-level compared to when given limited access to straw. Through hormonal regulation, this will have a positive effect on the farrowing process resulting in fewer stillbirths; and on the maternal behaviour reducing the number of piglets dying due to crushing. We also hypothesised that a more suitable micro-climate in pens with large amounts of straw will reduce the risk for hypothermia, resulting in less weak piglets dying from starvation. Altogether, we hypothesised that strategic use of straw will reduce the overall pre-weaning mortality in liveborn piglets. The objectives of the present study were therefore to quantify the effect of a large quantity of straw (15–20 kg) given before farrowing, compared to small daily amounts (0.5–1 kg), on:

(a) the number of stillborn piglets; (b) post-mortem findings in piglets dying within 5 days after birth; (c) the pre-weaning mortality in liveborn piglets.

2. Materials and methods This study was approved by the Regional Ethics Committee for animal experiments in Gothenburg. 2.1. Farms, housing and management The study was carried out in 2009 as four cohort trials on commercial piglet-producing farms (A, B, C and D) in SouthWest Sweden (Table 1). The farms were selected based on the following criteria: (a) farm situated within 50 km from the university campus in Skara; (b) liquid manure system capable of managing large quantities of straw; (c) piglet production based on batch-wise farrowing with at least 30 sows farrowing in the same batch; (d) all sows kept loose housed during farrowing and lactation; and (e) farmer and stockpersons willing to participate in the study. During gestation, sows were group-housed in deeplitter straw systems with individual feeding stalls. Sows entered the farrowing unit 5–4 days before the first sow in the batch was expected to give birth. In the farrowing unit, all sows were loose housed in pens with solid concrete flooring in 50% of the total pen area and 50% slatted flooring. On all four farms, creep areas were open with floor heating and a heating lamp. The slurry systems were based on liquid manure with mechanically operated scrapers directly under the slatted floor. Feeding and management were in accordance with the farms’ regular routines. Sows were routinely vaccinated against enterotoxigenic Eschericha coli, Erysipelothrix rhusiopathiae and porcine parvovirus. Farrowing occurred spontaneously, i.e. without pharmaceutical induction. Male piglets were castrated within 7 days of age. Tail docking was not performed. The stockpersons were allowed to apply cross-fostering of piglets between litters within the same treatment group. All crossfostered piglets were individually marked by making a small cut in one ear. All piglets were offered commercial piglet creep feed without antibiotics. Piglets were weaned at an average age of 5 weeks in accordance with current Swedish practice. 2.2. Treatments Researchers assigned pens to treatments before the sows entered the farrowing unit during the first study period. Every second pen was allocated to receive a large amount of straw (STRAW) while the adjacent pens were allocated as controls, receiving small amounts (CONTROL). The farmer was then asked to distribute the sows across pens, without knowledge about pen assignment. Hence treatment was randomised and in each farrowing unit the sows were equally distributed between treatments. On each farm, 54–68 sows were studied during one or two consecutive lactations (Study period 1: farrowings between March 10 and June 6; Study period 2: farrowings between August 10 and November 3). One hundred and twenty sows remaining in the batch during the second lactation switched treatment group. In total, 181 litters with access to large amounts of straw at birth (STRAW) and 182

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model

ARTICLE IN PRESS

PREVET-3761; No. of Pages 12

R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

3

Table 1 Overview of animals and housing conditions on participating farms.

Breeds of sows No. of sows in herd No. of sows entering study No. of litters entering study Size of farrowing pen Available free space for the sow Walls with farrowing rails Type of solid floor Type of slatted floor Slat width Width and length of slot openings

Farm A

Farm B

Farm C

Farm D

Swedish Landrace × Yorkshire 600 67 103 6.4 m2 (2.0 × 3.2 m) 4.6 m2 1 Concrete Cast iron 11 mm 11 × 200 mm

Swedish Landrace × Yorkshire 540 56 90 6.2 m2 (2.08 × 2.97 m) 4.5 m2 1 Concrete Cast iron 15 mm 10 × 200 mm

Swedish Landrace

Swedish Landrace × Yorkshire 594 68 96 6.2 m2 (1.95 × 3.2 m) 4.6 m2 1 Concrete Cast iron 11 mm 10 × 200 mm

= 15-20 kg

STRAW

= 2 kg

= 0.5-1 kg

etc.

X -2

Day 0

+5

Time (days)

CONTROL etc.

X Day 0

Time (days)

Fig. 1. Timeline of straw supply before expected farrowing (indicated by X) and after farrowing had occurred in treatments STRAW and CONTROL.

litters with small amounts of straw (CONTROL) entered the study. 2.3. Provision of straw STRAW sows were provided with 15–20 kg of chopped straw at 2 days prior to expected date of farrowing (Fig. 1). The straw volume corresponding to a weight of 15 kg was measured by markings in the farms’ straw carts. The aim was to keep the floor of the pen, including the piglet corner, completely covered with straw bedding until at least 1 day after farrowing. If the straw drained too quickly through the slatted floor, making more than a third of the solid or slatted flooring visible before farrowing, the farmers were instructed to provide additional straw to cover visible areas. Otherwise no additional straw was provided and the pens were not cleaned until 5 days after farrowing (day +5; Fig. 1). If any straw remained in the pen on day 5, it was removed manually. From this day onwards the pen was cleaned and 0.5–1 kg of chopped straw was given daily according to the farms’ regular management practices. This small amount of straw had usually disappeared by the next day. CONTROL sows received 0.5–1 kg of chopped straw on a daily basis, plus about 2 kg extra to be used for nestbuilding when the stockperson judged the sow to be about to farrow (Fig. 1). Straw was also put in the piglet corner. At the onset of farrowing, the stockpersons put an

266 54 74 6.0 m2 (1.97 × 3.07 m) 4.9 m2 3 Concrete Plastic 18 mm 11 × 95 mm

additional 1–2 kg of chopped straw behind the sow and around the piglets for extra comfort according to the farms’ regular management routines. 2.4. Recordings Sow identity, parity, expected farrowing date, actual farrowing date, number of piglets born alive, number of stillbirths, and number of piglets cross-fostered were recorded for each litter. Liveborn piglets were individually weighed within 36 h of birth (day 0 or 1) and the mean birth weight within the litter was calculated. The weights of piglets dying before the litter had been weighed were also included in the birth weight calculation. Piglets still alive five days after birth and at weaning were counted by the researchers. One litter was excluded at farrowing since the number of stillborn piglets was not recorded. At day 5, four litters were excluded because all original piglets had been cross-fostered to other sows. At weaning, eight sows were no longer present for various reasons (use of foster sow system, diseases etc.) and their litters where therefore excluded in the last counting, leaving 350 complete litters. The rectal body temperature of each sow was measured in the morning (between 7:00 and 11:00 am) on day 1 and day 2 post-partum. Body temperature was later categorised as a binary variable (0: <39.5 ◦ C on both occasions; 1: ≥39.5 ◦ C on day 1 and/or 2). Back-fat thickness was measured when the sows entered the farrowing unit, using ultrasound equipment (Krautkramer USM 22). The probe was placed at the level of the last rib, 6–7 cm to the side of the spinal column (processus spinosus) on each side. The mean of these two measurements was used. One researcher did all back-fat measurements. All collected data were manually transferred from recording sheets to Microsoft Excel 2007 (Microsoft Corp., Redmond, WA, USA) for logical checks. Table 2 shows the distribution of observations across levels of categorical independent variables. 2.5. Post-mortem examination All liveborn piglets that died within 5 days after birth were collected by the staff. Current date and sow identity

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model

ARTICLE IN PRESS

PREVET-3761; No. of Pages 12

R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

4

Table 2 Distribution of observations across levels of categorical independent variables in treatments STRAW and CONTROL; 362 litters. Variable

Level

STRAW

Farm

A B C D

52 45 36 48

29 25 20 27

51 45 38 47

28 25 21 26

Seasona

Spring Autumn

93 88

51 49

91 90

50 50

Parity

1 2 3 4 5 6–7 Missing

36 28 23 31 37 25 1

20 15 13 17 20 14 –

27 29 26 31 39 29 –

15 16 14 17 22 16 –

Sow body temperatureb , ◦ C

<39.5 ≥39.5 Missing

133 43 5

76 24 –

129 43 9

75 25 –

Mean birth weightc , g

<1400 1400–1599 1600–1799 ≥1800 Missing

41 53 45 41 1

23 29 25 23 –

42 53 51 32 3

23 30 29 18 –

No. of litters

a b c

CONTROL %

No. of litters

%

Spring: farrowing between March 10 and June 6; Autumn: farrowing between August 10 and November 3. Maximum temperature of recordings day 1 and day 2 after farrowing. Recorded within 36 h after birth.

was written with a marker pen on the piglet’s skin. In the afternoon the same day, a researcher transported the dead piglets to the Eurofins laboratory in Skara, where they were stored in a cold-storage room until post-mortem examination was performed. In most cases, the examination was made on the following day or at most within 3 days. All post-mortem examinations were performed by one of two pathologists to whom the history of the individual piglet was unknown. Diagnosis was based on macroscopic lesions only. Piglets were assigned to one of the four primary diagnoses; starvation, crushing, enteritis and miscellaneous. The diagnosis starvation was assigned if the piglet was emaciated and/or if the stomach and intestines were empty. Crushing was diagnosed if internal or external lacerations with bleedings were present. Enteritis included flaccid and dilated small intestine, congested mesentery vessels and/or fluid content of the large intestine. Piglets with other findings (congenital deformation, pneumonia, sepsis, etc.) or with no specific lesions were assigned miscellaneous. If a piglet was both starved and crushed, starvation was regarded as predisposing for crushing and the piglet was assigned starvation as primary diagnosis. All post-mortem findings (free text format) and the weight of each piglet was recorded. Stillborn piglets were recorded by the farmer as “stillborn” if they had intact “slippers” (cartilaginous tips) on their claws, but no laboratory examination was performed.

XTMEPOISSON procedure in Stata/IC® 11 (StataCorp LP, College Station, Texas, USA). The dependent variable was the number of stillborn piglets in the litter, and the natural logarithm of the number of piglets born in total (live + stillborn i.e. the number of individuals at risk) was used as offset. The effect of sow was tested initially and forced into the model. The following independent variables were then tested as fixed effects in univariable models: treatment (CONTROL; STRAW), farm (A; B; C; D), season, (spring; autumn), parity (1; 2; 3; 4; 5; ≥6) and back-fat thickness (in mm; centred around overall mean; polynomial extension to the third degree). Variables with p ≤ 0.25 were selected for further modelling. A multivariable model was constructed using manual backward stepwise elimination and retaining variables with p ≤ 0.05. Finally, previously excluded variables were forced into the model again and retained if p ≤ 0.05 or if they were judged to confound treatment (i.e. changed its estimate by >15%). One-way interactions of treatment with all other independent variables in the final model were tested but none was found significant (p > 0.05). Coefficient estimates were transformed into incidence rate ratios (IRR). The fit of the final model was evaluated through visual inspection of residuals plots and examination of outliers.

2.6. Statistical analysis of stillbirths

2.7. Statistical analysis of deaths due to starvation, crushing and enteritis

For statistical analysis of stillbirths, a mixed effects Poisson regression model was used, applying the

Three mixed effects Poisson regression models were used to analyse the number of piglets per litter dying

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

due to starvation, crushing and enteritis, again applying the XTMEPOISSON procedure in Stata. Piglets diagnosed with both starvation and crushing were only considered as “cases” in the starvation model. Log litter size after crossfostering was set as offset variable. The independent variables treatment, farm, season, parity, back-fat thickness (polynomial extension to the third degree), sow body temperature (0: <39.5 ◦ C; 1:≥39.5 ◦ C) and mean birth weight (0: <1400 g; 1: 1400–1599 g; 2: 1600–1799 g; 3: ≥1800 g) were tested as fixed effects in univariable models, and later offered to multivariable modelling if found significant at p ≤ 0.25. Manual backward stepwise elimination was used and variables with p ≤ 0.05 were retained. The variable farm and the random effect of sow were forced into all models. The final models were validated through residual plots and examination of outliers. Average predicted margins and their standard errors were calculated at the two levels of treatment. 2.8. Statistical analysis of pre-weaning mortality The effect of treatment on the number of liveborn pigs per litter dying until weaning was analysed using a Poisson regression model. Mortality rate was calculated as the number of liveborn piglets dying divided by the litter size after adjustment for cross-fostering; hence log litter size after cross-fostering was again set as offset variable. Initially, the random effect of sow was tested with the XTMEPOISSON procedure in Stata and found nonsignificant (p = 1.0). Therefore, the ordinary Poisson model was applied using the POISSON procedure, adjusted for within-sow correlation. Multivariable model building and validation was performed and in the same way as described for deaths due to starvation, crushing and enteritis. 3. Results 3.1. Number of stillbirths In the 362 studied farrowings, a total of 5247 piglets were born. Descriptive statistics of litter and sow traits related to treatment are shown in Table 3. Stillbirths accounted for 6.5% of all piglets born (340 piglets) and occurred in 49.7% of the litters. The mean incidence rate (±SD) of stillbirths in STRAW litters was 5.2 ± 8.2 per 100 born piglets (n = 181 litters), compared to 7.2 ± 10.4 stillbirths per 100 born piglets in CONTROL (n = 181 litters). In the multivariable model, the effect of treatment reduced stillbirths with 27% in STRAW litters (p = 0.007; Table 4). For the fixed portion, the model predicted a count of 0.61 stillborn piglets in STRAW compared to 0.84 in CONTROL litters. The number of stillborn was also affected by farm and parity, with an increasing number of stillborn piglets in older sows. Parity 5 sows had almost twice as many stillbirths with a 96% increase compared to parity 1 sows. 3.2. Timing and causes of deaths Post-mortem examination was performed in 798 piglets that died within 5 days after birth. Out of these, 3 were stillborn, 14 had non-readable identities and 9 came from the

5

four litters that were omitted, leaving 772 piglets in the analysis. The majority of deaths (77%) occurred within 72 h after birth. The three major causes of death were starvation (34%; n = 260), crushing (28%; n = 220), and enteritis (24%; n = 182). No specific lesions were found on 6.6% of the piglets while other conditions, including congenital abnormalities and pneumonia, represented 7.6% of mortality occurring within 5 days post-partum. Among starved piglets, 22% (n = 56) also showed signs of crushing which accounted for 7% of all examined piglets. Two piglets were starving and had enteritis and two others were both crushed and diagnosed with enteritis. In these cases, starvation and crushing were assigned as the primary causes of death. The distribution of post-mortem findings differed between the treatments. Among STRAW piglets examined (n = 405), the main diagnosis was crushing (34%) followed by enteritis (27%) and starvation (24%). In CONTROL litters nearly half of the examined piglets (n = 367) were starving (45%) and less piglets were crushed (23%) or suffered from enteritis (20%). Deaths due to crushing and enteritis peaked on day 1 in both treatments, while deaths due to starvation were most common on day 2 in the CONTROL and day 1 in the STRAW treatment (Fig. 2). The mean piglet birth weight (± SD) within a litter was 1573 ± 264 g (n = 358 litters). The mean birth weight of piglets that died within 5 days of age was 1241 ± 409 g. The smallest piglet weighed 400 g. At postmortem examination, the mean weight of dead piglets was 1198 ± 448 g. Piglets diagnosed with starvation weighed 876 ± 312 g while piglets dying from crushing were somewhat heavier, weighing 1292 ± 410 g. Piglets with enteritis weighed 1431 ± 389 g. 3.3. Effect of STRAW on deaths due to starvation, crushing and enteritis Treatment had a significant influence on both starved and crushed piglets (Tables 5 and 6). The mean incidence rates (± SD) of piglets starving to death until day 5 were 3.6 ± 6.0 and 6.3 ± 7.9 per 100 piglets in STRAW and CONTROL litters. The multivariable model predicted a 40% reduction with a predicted count of 0.89 starved piglets in CONTROL decreasing to 0.54 piglets in STRAW litters (p < 0.001). Losses due to starvation were also influenced by mean birth weight and parity, with 2.9–3.7 times more starved piglets in litters from multiparous sows compared to gilts. The number of starved piglets was reduced by 88% in litters with a mean birth weight ≥1800 g compared to <1400 g. There was no significant effect of farm, season, sow body temperature or back-fat thickness. The mean incidence rate of crushed piglets was 4.9 ± 7.6 per 100 piglets in STRAW and 3.2 ± 5.3 per 100 piglets in CONTROL litters. Piglet mortality due to crushing was affected by treatment, season and parity. Deaths due to crushing increased by 66% in STRAW litters (p = 0.001; Table 6) and the model predicted the number of crushed piglets to increase from 0.38 to 0.63 piglets per litter. Losses due to crushing increased in older sows (parity ≥5). These had 1.9–2.7 times more piglets dying from crushing compared to gilts.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

6

Table 3 Descriptive statistics of sow and litter traits in treatments STRAW and CONTROL. Trait

STRAW

Total number of born/litter Number of live-born/litter Number of stillbirths/litter Litter size after cross-fostering Litter size on day 5 Litter size at weaning Pre-weaning mortality, % Back-fat thickness at farrowing, mm

CONTROL

n

Mean ± SD

181 181 181 180 180 177 177 178

14.2 13.4 0.8 13.6 11.2 10.7 18.8 16.1

± ± ± ± ± ± ± ±

3.8 3.7 1.1 2.6 1.6 1.6 15.4 3.8

(Range)

n

Mean ± SD

(4–23) (3–21) (0–7) (8–23) (6–15) (6–14) (0–70) (8–27)

181 181 181 178 178 173 173 180

14.8 13.7 1.1 13.5 11.3 10.5 20.2 16.0

± ± ± ± ± ± ± ±

3.4 3.3 1.7 2.1 1.6 1.6 14.0 3.8

(Range) (2–22) (2–22) (0–11) (8–20) (6–15) (6–15) (0–61) (6–27)

Table 4 Mixed effects Poisson regression modela of number of stillbirths in 362 litters in 4 farms. Variable

Level

Coefficient

SE

IRRb

95% CI for IRR

Treatment

CONTROL STRAW

Baseline −0.32

– 0.12

1.00 0.73

– 0.57, 0.92

– 0.007

Farm

A B C D

Baseline 0.31 0.92 0.46

– 0.23 0.22 0.22

1.00 1.36 2.50 1.59

– 0.87, 2.14 1.61, 3.88 1.03, 2.45

– 0.18 <0.001 0.035

Parity

1 2 3 4 5 ≥6

Baseline 0.13 0.09 0.49 0.67 0.47

– 0.27 0.27 0.25 0.24 0.26

1.00 1.14 1.10 1.63 1.96 1.59

– 0.67, 1.92 0.64, 1.88 0.99, 2.67 1.23, 3.12 0.95, 2.66

– 0.63 0.73 0.053 0.005 0.075

−3.65

0.26

Intercept a b

p-value

<0.001

Sow included as random effect in the model, and the natural logarithm of the total number of piglets born used as offset. Incidence rate ratio.

The number of piglets diagnosed with enteritis was affected by treatment and farm, with STRAW increasing the predicted count from 0.26 to 0.39 piglets per litter (IRR = 1.52; p = 0.012). 3.4. Pre-weaning mortality Piglets were weaned at 32 ± 3 days of age (mean ± SD, n = 350 litters). The mortality of liveborn piglets until weaning was 19.5 ± 14.7%. In 56 litters (16%), no deaths occurred

until weaning; however in 118 litters (34%) mortality rates were extremely high (≥25%). There was a numerical difference between treatments with 18.8 ± 15.4% mortality in STRAW versus 20.2 ± 14.0% in CONTROL litters, but the effect was not statistically significant (Table 7). Preweaning mortality was however influenced by farm, parity and mean birth weight. Mortality increased gradually with higher parities and was almost twice as high in parities ≥6 as in parity 1 (IRR = 1.84, p < 0.001). Mortality decreased with higher mean birth weights. In a litter with a mean birth

Fig. 2. Mortality rate in live-born piglets during 5 days post-partum due to crushing (—), enteritis (- - -), starvation (· – · –) and miscellaneous causes (– – –) in STRAW (left; n = 2428) and CONTROL-treatment (right; n = 2382), based on post-mortem findings in 772 piglets from 358 litters in 4 farms. Day 0 = day of farrowing.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

7

Table 5 Mixed effects Poisson regression modela of number of starved piglets until day 5 after farrowing in 358 litters in 4 farms. Variable

Level

Coefficient

SE

IRRb

95% CI for IRR

p-value

Treatment

CONTROL STRAW

Baseline −0.51

– 0.13

1.00 0.60

– 0.47, 0.78

– <0.001

Farm

A B C D

Baseline 0.25 −0.07 0.03

– 0.17 0.20 0.18

1.00 1.28 0.93 1.02

– 0.92, 1.79 0.63, 1.39 0.72, 1.46

– 0.14 0.74 0.88

Parity

1 2 3 4 5 ≥6

Baseline 1.05 1.15 1.24 1.31 1.10

– 0.32 0.31 0.30 0.28 0.30

1.00 2.85 3.17 3.45 3.72 3.00

– 1.53, 5.30 1.73, 5.81 1.94, 6.12 2.15, 6.42 1.68, 5.36

– 0.001 <0.001 <0.001 <0.001 <0.001

<1400 1400–1599 1600–1799 ≥1800

Baseline −0.46 −1.14 −2.15

– 0.14 0.19 0.35

1.00 0.63 0.32 0.12

– 0.48, 0.84 0.22, 0.46 0.06, 0.23

– 0.002 <0.001 <0.001

−3.27

0.29

Mean birth weightc , g Intercept a b c

<0.001

Sow included as random effect in the model, and the natural logarithm of litter size after cross-fostering used as offset. Incidence rate ratio. Recorded within 36 h after birth.

Table 6 Mixed effects Poisson regression modela of number of crushed piglets until day 5 after farrowing in 358 litters in 4 farms. Variable

Level

Coefficient

SE

IRRb

95% CI for IRR

Treatment

CONTROL STRAW

Baseline 0.51

– 0.15

1.00 1.66

– 1.24, 2.22

– 0.001

Farm

A B C D

Baseline −0.24 −0.11 0.08

– 0.23 0.25 0.22

1.00 0.78 0.90 1.08

– 0.49, 1.24 0.55, 1.46 0.71, 1.65

– 0.30 0.66 0.72

Seasonc

Spring Autumn

Baseline 0.51

– 0.16

1.00 1.67

– 1.23, 2.28

– 0.001

Parity

1 2 3 4 5 ≥6

Baseline 0.30 0.23 0.40 0.64 0.99

– 0.31 0.33 0.31 0.28 0.28

1.00 1.35 1.25 1.49 1.89 2.69

– 0.74, 2.46 0.66, 2.39 0.82, 2.72 1.09, 3.27 1.55, 4.67

– 0.33 0.49 0.19 0.023 <0.001

−4.82

0.38

Intercept a b c

p-value

<0.001

Sow included as random effect in the model, and the natural logarithm of litter size after cross-fostering used as offset. Incidence rate ratio. Spring: farrowing between March 10 and June 6; Autumn: farrowing between August 10 and November 3.

weight of ≥1800 g, mortality was reduced by 54% compared to litters with a mean birth weight of <1400 g (p < 0.001). 4. Discussion The aim of the present study was to study the effects of a strategic use of straw at farrowing on piglet survival. The results show that the number of stillbirths and distribution of post-mortem findings differed between the two treatments although the pre-weaning mortality of liveborn piglets was not affected. 4.1. Stillbirths The percentage of stillborn piglets observed in the present study was 6.5%. This is within range of what others

have found. Reported rates of stillbirth under commercial conditions vary between 4 and 12% (Strange et al., 2013; Vanderhaeghe et al., 2010; Cutler et al., 2006). An association between prolonged farrowing and an increased probability of stillbirth has been shown in several studies (Canario et al., 2006; Borges et al., 2005) and is explained by the greater risk of hypoxia due to occlusion or damage of the umbilical cord or placental detachment as farrowing progresses (van Dijk et al., 2006). However, there are also studies with lacking evidence for such an association (Vanderhaeghe et al., 2010; Lucia et al., 2002). Low oxytocin concentration is associated with prolonged parturitions in sows (Algers and Uvnäs-Moberg, 2007; Oliviero et al., 2008). It has recently been shown that plasma oxytocin concentrations in sows could be increased by providing abundant nesting materials before

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

8

Table 7 Poisson regression modela of pre-weaning mortality of liveborn piglets in 350 litters in 4 farms. Variable

Level

Coefficient

SEb

IRRc

95% CI for IRR

Treatment

CONTROL STRAW

Baseline −0.07

– 0.06

1.00 0.93

– 0.83, 1.05

– 0.24

Farm

A B C D

Baseline 0.25 0.14 0.31

– 0.09 0.10 0.08

1.00 1.29 1.15 1.36

– 1.07, 1.55 0.95, 1.40 1.15, 1.60

– 0.006 0.16 <0.001

Parity

1 2 3 4 5 ≥6

Baseline 0.22 0.27 0.44 0.46 0.61

– 0.13 0.15 0.12 0.11 0.11

1.00 1.25 1.31 1.55 1.59 1.84

– 0.96, 1.62 0.98, 1.75 1.22, 1.98 1.28, 1.97 1.47, 2.28

– 0.10 0.066 <0.001 <0.001 <0.001

<1400 1400–1599 1600–1799 ≥1700

Baseline −0.17 −0.40 −0.77

– 0.07 0.09 0.12

1.00 0.84 0.67 0.46

– 0.73, 0.97 0.56, 0.80 0.37, 0.58

– 0.021 <0.001 <0.001

–4.51

0.12

Mean birth weightd , g Intercept a b c d

p-value

<0.001

Litter size after cross-fostering used as offset. Robust standard errors, adjusted for sow clusters. Incidence rate ratio. Recorded within 36 h after birth.

parturition, possibly due to a reduction in potential stress through allowance of natural behaviour (Yun et al., 2013). Provision of 15–20 kg of straw 2 days before farrowing gives the sow an extended possibility to build a farrowing nest compared to when small daily amounts are given. We therefore hypothesised that this would lead to a reduced stress level which would positively affect the farrowing process and therefore reduce stillbirths. In a study of sow behaviour and farrowing duration, conducted on about one third of the farrowings studied in the present study, a strong association was found between the sum of time spent on nest-building pre-partum and the duration of farrowing regardless of treatment. Farrowing duration decreased with an increase of nest-building activity (Westin et al., 2015). It was also demonstrated that strategic use of straw increases the nest-building behaviour pre-partum. However, the numerical difference found in farrowing duration between treatments (287 min in STRAW versus 319 min in CONTROL) was not statistically significant. In the present study, where records from more than twice as many farrowings are included (362 farrowings in present study, 131 farrowings in behaviour study), a significant difference in number of stillborn piglets was found indicating that the birth process is affected by the amount of straw provided for nest building as hypothesised. However, in the present study stillbirths were diagnosed by the farmers (intact “slippers”), not by postmortem examination (lungs never breathed air). It is therefore possible that some of the “stillborn” piglets were live at birth but too weak to walk so that the “slippers” (cartilaginous tips) on their claws were still intact after death. Some piglets could therefore have been misclassified as “stillborn”. Viability is also closely related to the progress of farrowing, as is the number of stillborn piglets. Piglets born late in the birth order and piglets born after a long inter birth interval often have low viability scores and elevated

concentrations of lactate in the blood due to hypoxic stress (Rootwelt et al., 2012; Pedersen et al., 2011). So although it is possible that some piglets were misclassified, it seems likely that strategic use of straw will positively affect the farrowing process with fewer piglets at risk of being stillborn or weak at birth. The effect on duration of farrowing however still needs to be verified.

4.2. Starvation Starvation was the most common cause of death, diagnosed in 34% of examined piglets. The number of starving piglets was strongly reduced in STRAW compared to CONTROL pens. The ultimate cause of death is often only the final act in a chain of events and starvation is by many found to be associated to viability at birth (Edwards, 2002; Herpin et al., 1996; English and Smith, 1975). As clearly demonstrated by Herpin et al. (1996) and others, highly viable piglets reach the udder sooner and suckle more rapidly than piglets of low viability. Thus, less viable piglets are predisposed to suffer from starvation. In a study by Pedersen and co-workers (2006), low piglet activity was related to litters with high proportions of stillbirths. The authors observed a decreased percentage of piglets that suckled during 8 h after birth of the first piglet and suggest that the consequences of problematic births could be viewed as a continuum with stillbirth as the most extreme outcome and varying degrees of reduced viability and vitality as less extreme outcomes. One possible reason for the reduced number of piglets dying from starvation observed in the STRAW treatment in our study is therefore that piglets in these litters were more viable at birth. The fact that fewer stillbirths also were observed in the STRAW litters supports this hypothesis. Extended possibilities to build a nest thus seem beneficial both in order to prevent stillbirths and to increase viability at birth.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

Starvation and viability are also closely linked to hypothermia. The piglet body temperature drops by approximately 2 ◦ C within 20 min after birth. Then body temperature gradually rises to a normal value of 39 ◦ C, normally within 48 h. The extent of the drop and recovery period are however highly variable depending on environmental temperature, birth weight, viability and ingestion of colostrum (reviewed by Herpin et al., 2002). Piglets born on slatted floor have a greater decrease in rectal temperature during the first 30 min post-partum compared to piglets born on solid concrete (Pedersen et al., 2013). The good insulating properties of straw are well-known. Under semi-natural conditions, farrowing nests made of straw by free-ranging domestic sows have allowed piglets to withstand rather cold temperatures (Algers and Jensen, 1990). Under experimental conditions, moving a piglet from a concrete floor on to a 10 cm thick straw bedding at an ambiant temperature of 10 ◦ C had the same thermal effect as raising the ambient temperature to 18 ◦ C (Stephens, 1971). The goal of “strategic use of straw at farrowing” is to use such a large amount of straw so that straw bedding (approximately 5 cm deep) covers both the solid and the slatted floor area at farrowing. An insulating and warm micro-climate is thus provided for the newborn piglets independent of where the loose housed sow chooses to farrow. It is likely that this increases viability, thus helping the piglets to reach the udder faster. It possibly also reduces the number of weak piglets dying of hypothermia already at the birth site. A third possible explanation for the reduced number of starved piglets is that nest-building may affect the milk production positively. As most hormones involved in the progress of parturition (progesterone, prolactin, relaxin, oestrogens, glucocorticoids) are also involved in lactogenesis (Farmer and Quesnel, 2009), it has been speculated that alterations of the endocrine state of the sow during late pregnancy affecting the birth process might also affect the production of colostrum (Quesnel, 2011; Foisnet et al., 2010). There is however a great lack of knowledge within this field of research. In one study, sows with a low colostrum yield (<3 kg) had more stillborn piglets in their litters compared to other sows (Quesnel, 2011). Nest-building has been found to influence the endocrine state of the sow (Yun et al., 2013) and recent findings by Yun et al. (2014) suggests that nest-building improves sow metabolic status, and successful colostrum intake measured via neonatal piglet serum IgG and IgM concentrations during early lactation. They also suggest that use of sufficient nesting materials might lead to improved lactose synthesis through hormonal regulation, and thereby increase the overall colostrum yield. The results from the present study and earlier findings showing a higher growth rate in STRAW piglets (Westin et al., 2014), point in this direction. Sow body temperature ≥39.5 ◦ C at day 1 and 2 after parturition, as a clinical sign of the mastitis-metritis-agalactia syndrome (MMA), was not associated with increased death due to starvation as found by others (Pedersen et al., 2006). In the present study, medical treatment with non-steroidal anti-inflammatory drugs (NSAID) and/or antibiotics was however often initiated and if successful, deaths due to

9

starvation are likely to be prevented. The amount of sows with fever was similar in both STRAW and CONTROL. Thus, the effect of strategic use of straw resulting in a reduced number of starved piglets cannot be explained by less sows having fever in the STRAW treatment. As in many other studies, body weight at birth was strongly associated with starvation. Several studies indicate that low birth weight is not associated with low viability per se (Rootwelt et al., 2012, 2013; Herpin et al., 1996), but there are also studies showing that small piglets are born with higher lactate levels indicating that they more often suffer from hypoxia during delivery (Kammersgaard et al., 2011; Pedersen et al., 2011). Nevertheless, small piglets have a greater surface to body mass ratio, which result in greater heat loss at birth (Herpin et al., 2002). Consequently, the temperature drop is larger and small individuals therefore are prone to suffer from prolonged hypothermia. It is also well documented that small individuals take longer time to achieve their first suckle (Kammersgaard et al., 2011; Herpin et al., 1996) which predispose them to die from starvation as well as other causes. To facilitate recovery and prevent the hypothermia from being fatal, provision of a warm and suitable environment is therefore even more important when birth weight is low. 4.3. Crushing There is growing evidence showing that sows that are able to build farrowing nests properly will express more maternal behaviour thus being more caring towards their offspring after birth. In a study by Andersen et al. (2005), sows that did not crush any of their piglets within the first 4 days after farrowing, performed more nest-building and showed a more protective mothering style compared to sows that crushed two or more piglets. Non crusher sows responded sooner to piglet distress calls and had more nose-contact with the piglets during posture change. Similar results are shown by Yun et al. (2013) who found higher carefulness scores in loose housed sows provided with abundant nesting materials (chopped straw, sawdust, branches, shredded newspaper and ropes) compared to loose housed and crated sows without access to nestbuilding material. Therefore we hypothesised that the sows would not crush as many piglets in STRAW as in the CONTROL treatment. However, unexpectedly more piglets died from crushing in STRAW pens. One possible explanation is that the straw bedding present in STRAW pens may have reduced the piglets’ use of the creep area. To stimulate the piglets to use the creep area, it is important to keep this place warmer than the floor in the sow area (Andersen et al., 2007). Even with a suitable creep area it is very difficult to attract the piglets away from the sow during the first days after farrowing (Vasdal et al., 2010). With an insulating and soft straw bedding present in the whole pen, it is not likely that the creep area will attract the piglets and consequently they will spend more time out in the pen and at risk of being crushed. Another possible reason is poor pen design. Before lying down, sows group their piglets together by extensively rooting, scratching and turning around on the lying surface (Blackshaw and Hagelsø, 1990). Weber et al. (2009)

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

10

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

states that 5 m2 should be the minimum space needed for the unimpeded performance of such grouping behaviour. Although sows in our study were kept in larger pens (6.0–6.4 m2 ), the free space available for the sow was only 4.5–4.9 m2 . In addition, farrowing rails on the maximum three sides of the farrowing pen were only present on one farm. These rails are supposed to protect piglets from crushing and have been proven effective in a study by Andersen et al. (2007). Weber et al. (2009) did however not found a significant association between presence of farrowing rails and reduced mortality but the authors speculate that protection bars might be necessary to prevent piglet losses if the pen is too small which might be the case in our study. Considering the possibility that piglets in the STRAW pens eventually spent more time close to the sow, a poor pen design may have had a larger impact on piglet mortality due to crushing in STRAW than in CONTROL, even if STRAW sows possibly were more careful. Further research needs to be performed before the hypotheses about influence on maternal behaviour and reduced deaths due to crushing can be rejected. 4.4. Enteritis Although piglets in STRAW litters were born into a more suitable environment and were less likely to starve, the proportion of piglets dying due to enteritis was higher in STRAW than in CONTROL litters. We believe that fewer medical treatments in STRAW litters could explain the higher death rate due to enteritis in STRAW pens since 31% of individual piglets were medically treated for diarrhoea in STRAW compared to 37% in CONTROL litters (data not presented). Perhaps that the faeces were less visible when deep straw bedding was present and that medical treatment was therefore not initiated in time. The high prevalence of diarrhoea could also have influenced the number of crushed piglets since many piglets were not healthy and perhaps not as alert as otherwise. Since 2008, field experiences of a new diarrhoeic syndrome have been reported in several countries including Sweden, referred to by some authors as the New Neonatal Porcine Diarrhoea Syndrome (NNPDS) (Kongsted et al., 2013; Melin et al., 2010). No known pathogens or management factors causing the outbreaks has to our knowledge yet been identified. Mortality in affected piglets is in general low, but in one Danish herd 74% of dead piglets were diagnosed with enteritis suggested to be cases of NNPDS and the herd mortality of liveborn was 21% (Kongsted et al., 2014). The impact in individual farms can thus be very high. Because of the high frequency of deaths due to enteritis in spite of vaccination of sows against enterotoxigenic Escherichia coli and high level of medical treatment of affected individuals we also suspect that NNPDS was present on the farms in our study.

housed sows. In surveys conducted in Norway, Switzerland and the UK, piglet mortality ranged from 11 to 15% (KilBride et al., 2012; Weber et al., 2009; Andersen et al., 2007). Among 186 Swedish piglet producing farms taking part in the Swedish national sow recording scheme, the average mortality rate was 17.0% when the present study was conducted (Svenska Pig, 2009). We believe this is mainly caused by the extremely high prevalence of diarrhoea present at the farms resulting in the proportion of deaths caused by enteritis to be higher than normal. In our study 24% of the examined dead piglets were diagnosed with enteritis. This is almost twice as many as reported in a detailed investigation conducted by Svendsen et al. (1975) who continuously followed 17 Danish herds during 2 years. Cause of death was diagnosed by post-mortem examinations and 13% of the losses from birth to weaning were caused by diseases from the gastrointestinal tract. On the other hand, enteritis can be difficult to diagnose and different laboratories may use different criteria at post-mortem examination which could explain some of the difference. In the earlier mentioned survey by KilBride et al. (2012) only 4% of liveborn deaths were caused by diarrhoea, but in their study diagnosis was based on farmer observations only. No statistical difference in pre-weaning mortality was found between treatments as hypothesised. However, if the pen design is adjusted to better suit the behaviour of the sow and her litter when large amounts of straw bedding is present, it is possible that the number of crushed piglets can be reduced which in turn would influence the pre-weaning mortality in favour of the STRAW treatment. Also without such a high prevalence of diarrhoea present on the farms, the number of piglets dying due to enteritis would have been reduced. Nevertheless, from the farmer’s point of view the number of piglets weaned is of interest but also their growth and health during the suckling period. Considering the fact that piglets born in STRAW pens suffer less from claw and limb bruising and that they are heavier at weaning compared to CONTROL litters (Westin et al., 2014), the method for strategic use of straw at farrowing seems to be more beneficial.

5. Conclusion The objectives of the present study were to study the effect of strategic use of straw at farrowing on stillbirths, post-mortem findings within 5 days after birth and the preweaning mortality of liveborn piglets. We conclude that provision of 15–20 kg of straw 2 days prior to the calculated date of farrowing strongly affects stillbirths, reducing the number of stillborn piglets per litter by 27%. Under the conditions studied, the pre-weaning mortality of liveborn piglets was not affected by treatment; however the distribution of post-mortem findings differed with fewer piglets dying due to starvation and more due to crushing and enteritis in STRAW litters.

4.5. Pre-weaning mortality of liveborn piglets Conflicts of interest The pre-weaning mortality of liveborn piglets was 19.5% which is higher than what is reported by others for loose

The authors have no conflicts of interest to declare.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

Acknowledgements This research was funded by the Swedish Farmers’ Foundation for Agricultural Research (project no. H0750355). We wish to thank the farmers for kindly providing us access to their farms and to their staff for helping out with all practical issues. We also wish to thank Pernilla Nilsson at Avelspoolen for providing equipment and advice regarding back-fat measurements and Nils Lundeheim at the Swedish University of Agricultural Sciences for helpful discussions regarding the statistical analysis. References Algers, B., Jensen, P., 1990. Thermal microclimate in winter farrowing nests of free-ranging domestic pigs. Livest. Prod. Sci. 25 (1–2), 177–181. Algers, B., Uvnäs-Moberg, K., 2007. Maternal behavior in pigs. Horm. Behav. 52 (1), 78–85. Alonso-Spilsbury, M., Ramirez-Necoechea, R., Gonzlez-Lozano, M., MotaRojas, D., Trujillo-Ortega, M.E., 2007. Piglet survival in early lactation: a review. J. Anim. Vet. Adv. 6 (1), 78–86. Andersen, I.L., Berg, S., Bøe, K.E., 2005. Crushing of piglets by the mother sow (Sus scrofa)—purely accidental or a poor mother? Appl. Anim. Behav. Sci. 93 (3–4), 229–243. Andersen, I.L., Tajet, G.M., Haukvik, I.A., Kongsrud, S., Bøe, K.E., 2007. Relationship between postnatal piglet mortality, environmental factors and management around farrowing in herds with loose-housed, lactating sows. Acta Agric. Scand. Sect. A – Anim. Sci. 57 (1), 38–45. Blackshaw, J.K., Hagelsø, A.M., 1990. Getting-up and lying-down behaviours of loose-housed sows and social contacts between sows and piglets during Day 1 and Day 8 after parturition. Appl. Anim. Behav. Sci. 25 (1–2), 61–70. Borges, V.F., Bernardi, M.L., Bortolozzo, F.P., Wentz, I., 2005. Risk factors for stillbirth and foetal mummification in four Brazilian swine herds. Prev. Vet. Med. 70 (3–4), 165–176. Canario, L., Cantoni, E., Le Bihan, E., Caritez, J.C., Billon, Y., Bidanel, J.P., Foulley, J.L., 2006. Between-breed variability of stillbirth and its relationship with sow and piglet characteristics. J. Anim. Sci. 84 (12), 3185–3196. Cronin, G.M., Smith, J.A., 1992. Effects of accommodation type and straw bedding around parturition and during lactation on the behaviour of primiparous sows and survival and growth of piglets to weaning. Appl. Anim. Behav. Sci. 33 (2–3), 191–208. Cronin, G.M., van Amerongen, G., 1991. The effects of modifying the farrowing environment on sow behaviour and survival and growth of piglets. Appl. Anim. Behav. Sci. 30 (3–4), 287–298. Cutler, R.S., Fahy, V.A., Cronin, G.M., Spicer, E.M., 2006. Preweaning mortality. In: Straw, B.E., Zimmerman, J.J., D’Allaire, S.D., Taylor, D.J. (Eds.), Diseases of Swine. , Ninth ed. Blackwell Publishing Ltd., pp. 993– 1010. Edwards, S.A., 2002. Perinatal mortality in the pig: environmental or physiological solutions? Livest. Prod. Sci. 78 (1), 3–12. Edwards, S.A., Furniss, S.J., 1988. The effects of straw in crated farrowing systems on peripartal behaviour of sows and piglets. Br. Vet. J. 144 (2), 139–146. English, P.R., Smith, W.J., 1975. Some causes of death in neonatal piglets. Vet. Annu. 15, 95–104. Farmer, C., Quesnel, H., 2009. Nutritional, hormonal, and environmental effects on colostrum in sows. J. Anim. Sci. 87 (13 suppl.), 56–64. Foisnet, A., Farmer, C., David, C., Quesnel, H., 2010. Relationships between colostrum production by primiparous sows and sow physiology around parturition. J. Anim. Sci. 88 (5), 1672–1683. Herpin, P., Damon, M., Le Dividich, J., 2002. Development of thermoregulation and neonatal survival in pigs. Livest. Prod. Sci. 78 (1), 25–45. Herpin, P., Le Dividich, J., Hulin, J.C., Fillaut, M., De Marco, F., Bertin, R., 1996. Effects of the level of asphyxia during delivery on viability at birth and early postnatal vitality of newborn pigs. J. Anim. Sci. 74 (9), 2067–2075. Jarvis, S., Lawrence, A.B., McLean, K.A., Deans, L.A., Chirnside, J., Calvert, S.K., 1997. The effect of environment on behavioural activity, ACTH, (␤-endorphin and cortisol in pre-farrowing gilts. Anim. Sci. 65 (03), 465–472. Kammersgaard, T.S., Pedersen, L.J., Jørgensen, E., 2011. Hypothermia in neonatal piglets: interactions and causes of individual differences. J. Anim. Sci. 89 (7), 2073–2085.

11

KilBride, A.L., Mendl, M., Statham, P., Held, S., Harris, M., Cooper, S., Green, L.E., 2012. A cohort study of preweaning piglet mortality and farrowing accommodation on 112 commercial pig farms in England. Prev. Vet. Med. 104 (3–4), 281–291. Kirkden, R.D., Broom, D.M., Andersen, I.L., 2013. Invited review: piglet mortality: management solutions. J. Anim. Sci. 91 (7), 3361–3389. Kongsted, H., Jonach, B., Haugegaard, S., Angen, O., Jorsal, S., Kokotovic, B., Larsen, L., Jensen, T., Nielsen, J., 2013. Microbiological, pathological and histological findings in four Danish pig herds affected by a new neonatal diarrhea syndrome. BMC Vet. Res. 9 (1), 206. Kongsted, H., Stege, H., Toft, N., Nielsen, J., 2014. The effect of New Neonatal Porcine Diarrhea Syndrome (NNPDS) on average daily gain and mortality in 4 Danish pig herds. BMC Vet. Res. 10 (1), 1–7. Lucia Jr., T., Corrêa, M.N., Deschamps, J.C., Bianchi, I., Donin, M.A., Machado, A.C., Meincke, W., Matheus, J.E.M., 2002. Risk factors for stillbirths in two swine farms in the south of Brazil. Prev. Vet. Med. 53 (4), 285–292. Melin, L., Wallgren, P., Mattsson, S., Stampe, M., Löfstedt, M., 2010. Neonatal diarrhea in piglets from E. coli vaccinated sows in Sweden. In: D’Allaire, S., Friendship, R. (Eds.), Proceedings of 21st IPVS Congress. July 18–21, Vancouver, Canada, p. 290. Oliviero, C., Heinonen, M., Valros, A., Hälli, O., Peltoniemi, O.A.T., 2008. Effect of the environment on the physiology of the sow during late pregnancy, farrowing and early lactation. Anim. Reprod. Sci. 105 (3–4), 365–377. Pedersen, L.J., Berg, P., Jørgensen, G., Andersen, I.L., 2011. Neonatal piglet traits of importance for survival in crates and indoor pens. J. Anim. Sci. 89 (4), 1207–1218. Pedersen, L.J., Jørgensen, E., Heiskanen, T., Damm, B.I., 2006. Early piglet mortality in loose-housed sows related to sow and piglet behaviour and to the progress of parturition. Appl. Anim. Behav. Sci. 96 (3–4), 215–232. Pedersen, L.J., Malmkvist, J., Kammersgaard, T., Jørgensen, E., 2013. Avoiding hypothermia in neonatal pigs: effect of duration of floor heating at different room temperatures. J. Anim. Sci. 91 (1), 425–432. Quesnel, H., 2011. Colostrum production by sows: variability of colostrum yield and immunoglobulin G concentrations. Animal 5 (10), 1546–1553. Rootwelt, V., Reksen, O., Farstad, W., Framstad, T., 2012. Blood variables and body weight gain on the first day of life in crossbred pigs and importance for survival. J. Anim. Sci. 90 (4), 1134–1141. Rootwelt, V., Reksen, O., Farstad, W., Framstad, T., 2013. Postpartum deaths: piglet, placental, and umbilical characteristics. J. Anim. Sci. 91 (6), 2647–2656. Stephens, D.B., 1971. The metabolic rates of newborn pigs in relation to floor insulation and ambient temperatures. Anim. Prod. 13 (2), 303–313. Strange, T., Ask, B., Nielsen, B., 2013. Genetic parameters of the piglet mortality traits stillborn, weak at birth, starvation, crushing, and miscellaneous in crossbred pigs. J. Anim. Sci. 91 (4), 1562–1569. Svendsen, J., Bille, N., Nielsen, N.C., Larsen, J.L., Riising, H.J., 1975. Preweaning mortality in pigs, 4 Diseases of the gastrointestinal tract in pigs. Nord. Vet. Med. 27 (2), 85–101. Svenska Pig, 2009. PigWin Sugg Medeltal 2009, Official web page: www.pigwin.se/medeltal-sugg (date of entry 23-06-2014). Uvnäs-Moberg, K., Johansson, B., Lupoli, B., Svennersten-Sjaunja, K., 2001. Oxytocin facilitates behavioural, metabolic and physiological adaptations during lactation. Appl. Anim. Behav. Sci. 72 (3), 225– 234. van Dijk, A.J., van der Lende, T., Taverne, M.A.M., 2006. Acid–base balance of umbilical artery blood of liveborn piglets at birth and its relation with factors affecting delivery of individual piglets. Theriogenology 66 (8), 1824–1833. Vanderhaeghe, C., Dewulf, J., De Vliegher, S., Papadopoulos, G.A., de Kruif, A., Maes, D., 2010. Longitudinal field study to assess sow level risk factors associated with stillborn piglets. Anim. Reprod. Sci. 120 (1–4), 78–83. Vasdal, G., Glærum, M., Meliˇsová, M., Bøe, K.E., Broom, D.M., Andersen, I.L., 2010. Increasing the piglets’ use of the creep area—A battle against biology? Appl. Anim. Behav. Sci. 125 (3–4), 96–102. Weber, R., Keil, N.M., Fehr, M., Horat, R., 2009. Factors affecting piglet mortality in loose farrowing systems on commercial farms. Livest. Sci. 124 (1–3), 216–222. Westin, R., Holmgren, N., Hultgren, J., Algers, B., 2014. Large quantities of straw at farrowing prevents bruising and increases weight gain in piglets. Prev. Vet. Med. 115 (3–4), 181–190. Westin, R., Holmgren, N., Mattsson, B., Algers, B., 2013. Throughput capacity of large quantities of chopped straw in partly slatted farrowing

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023

G Model PREVET-3761; No. of Pages 12

12

ARTICLE IN PRESS R. Westin et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx

pens for loose housed sows. Acta Agric. Scand. Sect. A – Anim. Sci. 63 (1), 1–10. Westin, R., Hultgren, J., Algers, B., 2015. Strategic Use of Straw Increases Nest-Building in Loose Housed Farrowing Sows. Appl. Anim. Behav. Sci. (In Press). Yun, J., Swan, K.-M., Vienola, K., Farmer, C., Oliviero, C., Peltoniemi, O., Valros, A., 2013. Nest-building in sows: effects of farrowing housing on

hormonal modulation of maternal characteristics. Appl. Anim. Behav. Sci. 148 (1–2), 77–84. Yun, J., Swan, K.M., Vienola, K., Kim, Y.Y., Oliviero, C., Peltoniemi, O.A.T., Valros, A., 2014. Farrowing environment has an impact on sow metabolic status and piglet colostrum intake in early lactation. Livest. Sci. 163 (0), 120–125.

Please cite this article in press as: Westin, R., et al., Post-mortem findings and piglet mortality in relation to strategic use of straw at farrowing. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.02.023