Risk factors for development of lameness in gestating sows within the first days after moving to group housing

The Veterinary Journal 220 (2017) 28–33

Contents lists available at ScienceDirect

The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l

Risk factors for development of lameness in gestating sows within the first days after moving to group housing Liesbet M. Pluym a,b, Dominiek Maes b, Stephanie Van Weyenberg a, Annelies Van Nuffel a,* a Technology and Food Sciences Unit, Institute of Agricultural and Fisheries Research, Burgemeester Van Gansberghelaan 115 bus 1, Merelbeke 9820, Belgium b Department of Reproduction, Obstetrics and Herd Health, Ghent University, Salisburylaan 133, Merelbeke 9820, Belgium

A R T I C L E

I N F O

Article history: Accepted 16 November 2016 Keywords: Sow Lameness Risk factors Group housing Floor area

A B S T R A C T

Lameness in sows is an important welfare issue that is affected by housing conditions and is thought to be influenced by hierarchical fights within the first days after mixing sows in groups. A longitudinal study in 15 randomly selected herds was performed to investigate the incidence of sow lameness and possible risk factors within the first days of group housing. Each herd was visited just before and again 3–5 days after the sows were moved to group housing. The floor characteristics and dimensions of the group housing facilities were assessed. Locomotion ability, body condition, skin lesions and degree of faecal soiling were recorded for all sows. Additional information on housing and management was obtained using a questionnaire. Amongst the 810 sows included in the study, the mean lameness incidence was 13.1% (95% confidence interval 10.9–15.6%). Following binomial logistic regression analysis, sows with >10% of the body covered with faeces had an increased risk for development of lameness (odds ratio, OR = 2.33, P = 0.001). An increase in space allowance from 1.7 m2 to 3.0 m2 (OR = 0.40, P = 0.03) and of herd size from 144 to 750 sows per herd (OR = 0.71, P = 0.02) decreased the risk of development of lameness. Neither the degree of aggression, indicated by skin lesions, nor the floor characteristics influenced the development of lameness. These results indicate that sows can benefit from a larger floor area. © 2016 Published by Elsevier Ltd.

Introduction Group housing of gestating sows has been mandatory in all member states of the European Union since 2013.1 Although the change in sow housing was primarily driven by welfare concerns (Appleby, 2005), group housing may also present welfare issues, including injuries caused by post-mixing aggression and a higher prevalence of lameness (Gjein and Larssen, 1995a; Anil et al., 2003; Estienne et al., 2006; Chapinal et al., 2010). Lameness occurs in 8–27% of group housed sows (Bonde et al., 2004; Heinonen et al., 2006; KilBride et al., 2009; Pluym et al., 2011; Cador et al., 2014), although the number of lame sows can change throughout the reproductive cycle (Pluym et al., 2013) and during the period of group housing (Kroneman et al., 1993; Gjein and Larssen, 1995b; Calderón Díaz et al., 2013; Knox et al., 2014). Most of the lameness cases during group housing develop shortly after introduction of sows into the group (Kroneman et al., 1993; Anil et al., 2005; Chapinal et al.,

* Corresponding author. E-mail address: [email protected] (A. Van Nuffel). 1 See: European Commission, 2008. Council Directive 2008/120/CE of 18 December 2008 Laying Down Minimum Standards for the Protection of Pigs. http:// eur-lex.europa.eu/legal-content/EN/TXT/?qid=1474487225453&uri=CELEX: 32008L0120 (accessed 21 September 2016). http://dx.doi.org/10.1016/j.tvjl.2016.11.008 1090-0233/© 2016 Published by Elsevier Ltd.

2010; Knox et al., 2014). Kroneman et al. (1993) reported an incidence of lameness of 10% within the first month of mixing. However, there is a lack of more recent data on the incidence of lameness shortly after transferring sows to group housing. Lameness is influenced by housing conditions, including floor space allowance, group size and flooring. The impact of space allowance on the development of lameness in group housed sows has been studied, but the results are inconsistent (Gjein and Larssen, 1995b; Heinonen et al., 2006; Salak-Johnson et al., 2007; Willgert et al., 2014). The effect of space allowance on development of lameness may be dependent on group size, as demonstrated for finishing pigs (Street and Gonyou, 2008). However, studies investigating the association of group size and space allowance with development of lameness at sow level are lacking for group housed sows. Bare, slatted concrete floors, which are predominantly used for sow group housing, have been associated with development of lameness (Andersen and Bøe, 1999; Heinonen et al., 2006; KilBride et al., 2009). Slipperiness, abrasiveness, hardness, surface profile, void ratio and cleanliness are the main characteristics contributing to the injury potential of a floor (Webb and Nilsson, 1983; Webb, 1984; McKee and Dumelow, 1995). However, there has been limited investigation of these characteristics as risk factors for sow lameness (Cador et al., 2014). Furthermore, standards for floor characteristics, other than slipperiness, are lacking (Penny et al., 1965; Thorup et al., 2007).

L.M. Pluym et al. / The Veterinary Journal 220 (2017) 28–33

29

Table 1 Herd size, number of sows included in the study, sow breed and feeding system during gestation for each of the 15 herds in the study. Herd identification 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 a

Herd size

Sows included in study

Sow breed

Feeding system

350 560 210 450 180 210 144 750 550 240 225 400 745 450 500

35 98 30 55 22 37 30 60 61 23 9 89 98 72 91

Finnish Landrace Dalland Topigs 20 Topigs 20 Topigs 20 Danbred John Sykes Rymer (JSR) Genetics Danbred Pig Improvement Company (PIC) Crossbreds Belgian Landrace JSR Genetics Dalland Topigs 20 PIC

Free access stalls Free access stalls Free access stalls Trough feeding, no barriers Free access stalls Free access stalls Electronic sow feeders Free access stalls Trough feeding, partial barriers Free access stalls Ad libitum feeding Electronic sow feeders Trough feeding, no barriers Free access stalls Vario-Mixa

Feeding system without identification, with one feeding place and a storage space for dry feed.

In sows, social ranking is established after 2–3 days (Arey and Edwards, 1998). Within the first few hours following grouping, aggression may be intense and can result in skin lesions and other injuries (Turner et al., 2006). Aggressive encounters amongst sows have been suggested to result in lameness, but the association has not been confirmed (Kroneman et al., 1993; Gjein and Larssen, 1995b; Chapinal et al., 2010). In the present study, the incidence of lameness within the first 3–5 days of group housing was determined, with the aim to identify herd and sow level risk factors for development of lameness. Materials and methods Study design and study population Data were collected from 15 herds, randomly selected using the pig herd national database of the Belgian Federal Agency for Food Safety, during a longitudinal study from August 2012 to May 2013 (Table 1). The following eligibility criteria were applied: (1) sow herd or farrow-to-finish herd; (2) group housing of gestating sows;

(3) use of a batch production system; and (4) willingness of the farmer to participate. Herds using bedding material during group housing were excluded. The median herd size was 400 sows (range 144–750 sows). One batch of sows (9–99 sows) from each of the 15 herds was randomly selected for the study.

Data collection Four weeks after artificial insemination, the sows were moved from the insemination unit (i.e. individual stalls) to the gestation unit (i.e. group housing). Herds were visited just before and again 3–5 days after the sows were moved to group housing. Data collection comprised assessment of flooring and group size in the gestation unit, as well as locomotion, body condition, skin lesions and scoring for faecal soiling.

Flooring and group size Flooring ‘dirtiness’, quality, wetness and slip resistance at the gestation unit were assessed during the second herd visit according to a standardised protocol (Table 2). The mean floor area available to each sow and the percentages of slatted and solid floor were calculated. Group size was determined as the number of animals per pen.

Table 2 Overview of the protocol to assess the floor and sow characteristics measured as possible risk factors for development of lameness in sows. Risk factor

Method

Dirtiness

Visual assessment

Quality

Visual assessment

Wetness

Hygrometer (HM8-BF30, Merlin Technology GmbH)

Slip resistance

Portable Skid Resistance Tester (Munro Instruments) with TRL rubber slidera Renco Lean Meter (Renco Corporation)

Body condition Skin lesions Sow dirtiness

a b

Visual assessment Visual assessment (Welfare Quality, 2009)

Scoring system or unit Percentage of the floor covered with faeces Score 0: No faeces on the floor (clean) Score 1: 25% Score 2: 50% Score 3: 75–100% of the floor (severely dirty) Quality of the floor Score 0: Good quality flooring, without any enlarged gaps, protruding objects or level differences between successive slats Score 1: Presence of enlarged gap width Score 2: Presence of protruding sharp objects Score 3: Combination of score 1 and 2 Continuous (%)

Continuous (British pendulum number = coefficient of friction × 100) Backfat thickness (mm) Total number of skin lesions % of body soiled with faeces Score 0: <10% of body soiled (clean) Score 1: 10–30% of body soiled Score 2: >30% of body soiled (very dirty)

Location Entire stable

Entire stable

At least every third of the dunging area Free access stalls: front and rear half of 10% of stalls Other group housing systems: in two random locations of each lying area and around every feeder and drinking valve At least every third part of the dunging areab

P2 position (6–8 cm from dorsal mid-line at the level of the last rib) Whole body (excluding the tail) Both sides of body

At each herd, the Skid Resistance Tester was calibrated; TRL, Transport Research Laboratory. At each point the mean of eight readings was taken; the floor was not cleaned before measurements were performed.

30

L.M. Pluym et al. / The Veterinary Journal 220 (2017) 28–33

Table 3 Description (means ± standard deviation, SD) of continuous variables concerning development of lameness within the first 3–5 days of group housing in 810 group housed sows from 15 Belgian pig herds and results of the univariable analyses with development of lameness (yes/no) as dependent variable and herd as random effect. Variable

Total Mean ± SD

Sow related Backfat thickness Skin lesions Herd related Herd size (number of sows) Total area/sow (m2) Solid area/sow (m2) Flooring gestation unitb Wetness (%) Lying area Walking area

14.7 ± 3.7 14 ± 17 477 ± 178 2.3 ± 0.27 0.96 ± 0.48 88.0 ± 37.1 2.3 ± 0.91 2.6 ± 0.68

Lame

Non-lame

OR (95% CI)

P value

Range

Mean ± SD

n

Mean ± SD

na

7.0–30.0 0–115

15.3 ± 3.7 17 ± 20

106 106

14.6 ± 3.7 14 ± 16

704 704

1.05 (1.00–1.11) 1.01 (1.00–1.02)

0.058* 0.073*

144–750 1.67–3.0 0–1.65 29.0–147.0

– – – –

106 93 93 81

– – – –

704 628 628 507

0.72 (0.56–0.94) 0.23 (0.08–0.65) 0.53 (0.23–1.23)

0.015* 0.005* 0.138* 0.464

0.95–5.20 1.50–4.50

– –

81 92

– –

583 575

0.80 (0.59–1.07)

0.130* 0.463

a

a

Number of sows for which data were available. Slipperiness (British pendulum number). * Variables included in the multivariable logistic regression model. OR, odds ratio; CI, confidence interval. b

Lameness development Locomotion was assessed during the first and second herd visit. Sows had to walk until at least eight step cycles could be observed in a steady state walking pace. A three point numerical scale, adapted from Main et al. (2000), was used with score 0 (non-lame sow); score 1 (mildly lame sow; uneven weight distribution and stride length); and score 2 (severely lame sow; no weight bearing/lying down). The locomotion score for both herd visits was transformed into a binary ‘lameness development score’ by taking the non-lame sows into account from the first visit and further classifying these sows as ‘not lame’ in the case of score 0 on the second visit, or ‘became lame’ in the case of score 1 or 2 on the second visit. Body condition, skin lesions and sow ‘dirtiness’ Body condition was evaluated during the first herd visit. Sow faecal soiling (‘dirtiness’) and skin lesions were evaluated during the first and second visit. The protocol for assessing body condition, skin lesions and ‘dirtiness’ is given in Table 2; for each sow, the number of skin lesions on the body was counted and summed before entry into group housing and again during the second visit, from which the pre-grouping lesion score was subtracted to provide the number of lesions resulting from the postgrouping period (Geverink et al., 2009). At the first herd visit, the farmer was asked to fill in a questionnaire concerning general herd characteristics (herd size and type), housing (group housing system, flooring type and material), management (origin of gilts, feeding strategy at rearing, cleaning of the gestation unit) and feeding and water provision practices of the gestation unit. To control for information bias, the answers were checked by the first author in face-to-face interviews of the farmers and by inspection of the housing. Information on sow parity and breed were obtained from the farm records. Data management and statistical analysis Continuous variables with suboptimal distribution and categorical variables with <10% of observations in one category were either redefined or categorised (SPSS 21.0, IBM). Potentially explanatory variables used for statistical analysis are summarised in Tables 3 and 4. In order to examine the association between these variables and development of lameness, multilevel binary logistic regression analysis was performed using the PROC GLIMMIX procedure (SAS 9.3, SAS Institute). Model building started with univariable evaluation of all variables (Tables 3 and 4). To account for clustering of sows in the herd, a random herd effect was included. Only those variables associated with development of lameness (likelihood ratio chi-square test, P ≤ 0.25) were used to build the multivariable model (Hosmer and Lemeshow, 2000). Spearman’s rank correlations between the explanatory variables were tested. If the correlation was >0.6, only one of the two variables was selected. Since all the sows were clean during the first herd visit, only sow ‘dirtiness’ as scored during the second visit was included in the model. A manual stepwise backward model building procedure was followed until a model was obtained in which all the variables were significant at P < 0.05. All two-way interactions between significant variables were assessed. During the stepwise backward procedure, changes in the estimated coefficients of the variables were checked to evaluate for confounding. The fit of the model was assessed by inspection of the herd level standardised residuals plotted against the normal scores and against the herd level predicted values (Dohoo et al., 2009).

Results Descriptive statistics are summarised in Tables 3 and 4. All herds had solid concrete flooring in the lying area of the gestation unit.

Flooring in the slatted area for defaecation (‘dunging’ area) was made of concrete in 14 herds and metal slats in one herd. The median percentage of slatted floor in the gestation unit was 56% (range 32– 100%). In 12 herds, flooring in group housing was of good quality (score 0); in 53% of the herds, the floors were moderately to severely soiled (median ‘dirtiness’ score = 2). Grip of ‘dunging’ areas differed greatly between herds, with a mean (±standard deviation, SD) slip resistance of 88.0 ± 37.1 British pendulum number. In general, the lying areas were slightly drier than the ‘dunging’ areas, but for both areas wetness varied considerably amongst herds. A total of 810 eligible sows were included in the study; these sows had farrowed on average 2.7 (0–12) times and had a mean ± SD backfat thickness of 15 ± 3.7 mm. After 3–5 days of group housing, 32% of the sows were ≥10% covered with faeces and 73% had skin lesions. The frequency distribution of skin lesions had a positive skew (skewness = 2.0). The prevalence of ‘dirty’ sows (>10% of body soiled) varied between herds, with a median of 26% (range 8.1–64.3) in each herd. In total, 106 sows (13.1%, 95% confidence interval 10.9–15.6) developed lameness within the first 3–5 days of group housing and, 24/810 (3.0%) sows were severely lame. Parity, backfat thickness, skin lesions, sow ‘dirtiness’, herd size, type of group housing, origin of gilts, group size, total and solid area available per animal, wetness of lying area and cleaning were associated with lameness in the univariable analyses and were included in the multivariable logistic model (Tables 3 and 4). Due to the high correlation (ρ = 0.64; P < 0.001) between the variables ‘total area per sow’ and ‘wetness of lying area’, the latter was not included in the multivariable model. Table 5 shows the final multivariable logistic regression model. ‘Dirty’ sows with >10% of the body covered with faeces had significantly higher odds (OR = 2.33; P < 0.001) for development of lameness compared to clean sows. An increase in floor area per sow from 1.7 m2 to 3.0 m2 (OR = 0.40; P = 0.031) and in herd size from 144 to 750 sows (OR = 0.71; P = 0.020) decreased the odds for development of lameness. None of the two-way interactions between the remaining variables were significant (P > 0.05). No confounding factors were detected.

Discussion The mean lameness incidence (13.1%) in this study was slightly higher than the incidence (10%) observed by Kroneman et al. (1993). Kroneman et al. (1993) estimated the incidence over a period of 1 month, whereas only the first 3–5 days was included in the present study. Since the incidence of lameness may change during group housing (Kroneman et al., 1993; Knox et al., 2014), comparisons

L.M. Pluym et al. / The Veterinary Journal 220 (2017) 28–33

31

Table 4 Description of categorical variables related to development of lameness within the first 3–5 days of group housing in 810 group housed sows from 15 Belgian pig herds and results of the univariable analyses with development of lameness (yes/no) as dependent variable and herd as random effect. Variable Parity 0–1st parity 2nd parity 3rd–5th parity ≥6th parity Breeda Pig Improvement Company (PIC) Topigs 20 Dalland Danbred John Sykes Rymer (JSR) Genetics Otherb Sow dirtiness <10% dirty ≥10% dirty Herd type Farrow-to-finish herd Farrowing herd Group housing Free access stalls Electronic sow feeders Vario-mix and ad libitum Trough feeding Origin of gilts Rearing Purchasing Feeding strategy at rearing Ad libitum Restricted Group size ≤15 16–30 >30 Flooring of gestation unit Quality Good Badc Dirtiness (% soiled) Clean (0%) Slight (25%) Moderate (50%) Severe (≥75%) Cleaning of gestation unit Once per year Regular, dry Regular, wet Regular, wet + disinfection Water supply Ad libitum Restricted

n

Lame n (%)

Non-lame n (%)

OR (95% CI)

298 130 282 100

49 (46.2) 8 (7.5) 36 (34.0) 13 (12.3)

249 (35.4) 122 (17.3) 246 (34.9) 87 (12.4)

Reference 0.32 (0.15–0.71) 0.70 (0.40–1.22) 0.70 (0.33–1.47)

152 187 196 93 98 84

18 (17.0) 33 (31.1) 19 (17.9) 12 (11.3) 15 (14.2) 9 (8.5)

134 (19.0) 154 (21.9) 177 (25.1) 81 (11.5) 83 (11.8) 75 (10.7)

515 240

52 (57.1) 39 (42.8)

463 (69.7) 201 (30.3)

430 380

58 (54.7) 48 (45.3)

372 (52.8) 332 (47.2)

377 119 100 214

43 (40.6) 21 (19.8) 10 (9.4) 32 (30.2)

334 (47.7) 98 (13.9) 90 (12.8) 182 (25.9)

Reference 1.90 (0.86–4.17) 0.65 (0.18–2.36) 1.69 (0.90–3.17)

505 305

54 (50.9) 52 (49.1)

451 (64.1) 253 (35.9)

Reference 1.80 (1.15–2.81)

294 516

40 (37.7) 66 (62.3)

254 (36.1) 450 (63.9)

361 240 209

38 (35.8) 27 (25.5) 41 (38.7)

323 (45.9) 213 (30.3) 168 (23.9)

468 342

59 (55.7) 47 (44.3)

409 (58.1) 295 (41.9)

91 300 252 167

10 (9.4) 45 (42.5) 32 (30.2) 19 (17.9)

81 (11.5) 255 (36.2) 220 (31.3) 148 (21.0)

256 277 179 98

40 (37.7) 39 (36.8) 17 (16.0) 10 (9.4)

216 (30.7) 238 (33.8) 162 (23.0) 88 (12.5)

438 372

58 (54.7) 48 (45.3)

380 (54.0) 324 (46.0)

P value 0.043*

0.255

0.017* Reference 1.73 (1.10–2.70) 0.463

0.158*

0.010*

0.741

0.006* Reference 1.08 (0.64–1.82) 2.07 (1.28–3.35) 0.254

0.482

0.107* Reference 0.92 (0.48–1.80) 0.47 (0.22–1.00) 0.50 (0.22–1.15) 0.887

Breed represented by >60% of the sow population. Finnish Landrace, Belgian Landrace and crossbreds. c Presence of enlarged gaps and protruding objects. * Variables included in the multivariable logistic regression model. OR, odds ratio; CI, confidence interval. a

b

between these studies might not be possible. Moreover, according to the results of this study, the incidence of lameness differs between herds. Albeit restricted to the population of Belgian herds, the external validity of the present results is likely to be high, since it was

based on data from 15 herds, whereas the study by Kroneman et al. (1993) was based on data from only one herd. An increase in herd size from 144 sows to 750 sows lowered the odds of developing lameness. The percentage of sows that developed

Table 5 Final multivariable logistic regression model, with herd as a random effect, for risk factors for lameness in group housed sows from 15 randomly selected herds. Variable

Categories

Sows (n)

Herds (n)

Regression coefficient ± SE

OR (95% CI)

P value

Sow dirtiness

<10% dirty ≥10% dirty

515 240 810 721

14 14 15 14

Reference 0.85 (0.25) −0.34 (0.14) −0.93 (0.43)

2.33 (1.42–3.85) 0.71 (0.54–0.95) 0.40 (0.17–0.92)

0.001 0.020 0.031

Herd size (number of sows) Total floor area/sow (m2) SE, standard error; OR, odds ratio; CI, confidence interval.

32

L.M. Pluym et al. / The Veterinary Journal 220 (2017) 28–33

lameness decreased logarithmically with increasing herd size, without the presence of a clear threshold value. Herd size was significantly correlated with group size, but the correlation coefficient was relatively low (ρ = −0.38; P < 0.001). Group size was not retained in the final model, which suggests that group size in the present herds did not influence the relationship between herd size and development of lameness. The decreasing odds for lameness with increasing herd size may be explained by differences in management. A high correlation (ρ = −0.58; P < 0.001) was found between herd size and the origin of the breeding gilts, indicating that the owners of large herds reared their own gilts rather than purchasing new gilts. In the univariable analysis, the purchasing of gilts from breeding companies was significantly associated with higher odds for development of lameness compared to the selection of gilts from the same herd. The results of the present study suggest that rearing of breeding gilts from the same farm may promote the development of sound feet and legs. Sows also benefited from an increase in floor space allowance from 1.7 m2 to 3.0 m2. The minimal floor space requirements for gestating sows are regulated by European Union law. In one herd, multiparous sows were provided with floor spaces of 1.8 m2 instead of the legal minimum of 2.25 m2. However, the incidence of lameness in this herd was lower than the mean of 13.1%, which indicates that other factors had a stronger influence on lameness in this herd. Group size was not retained in the final model and no significant interaction was found with space allowance. This suggests that, in the present herds, the effect of increasing space allowance was independent of group size. Sows with >10% of the body covered with faeces had a higher risk of developing lameness within the first days of group housing compared to clean sows (<10% of body surface soiled). Zurbrigg and Blackwell (2006) suggested that there is an association between sow ‘dirtiness’ and lameness. In accordance with Geverink et al. (2009), the significant association between ‘dirtiness’ and lameness suggests that ‘dirtiness’ could be an indirect indicator of sow welfare. What caused the increased amount of faeces on the body of lame sows remains unclear. Since all the sows were scored clean during the first herd visit, sow dirtiness appears to have resulted from the type of housing in the gestation unit (i.e. group housing). Lame sows are less active, have shorter standing times and explore less compared to non-lame sows (Madec et al., 1986). It is possible that lame sows become dirtier because they lie down more often. It is also possible that, for lame sows, the cost of competing for a lying place is higher than lying in the ‘dunging’ area (Galindo and Broom, 2002). However, sow behaviour was not recorded in our study. The highest percentage of ‘dirty’ sows was found in the herds with the highest floor ‘dirtiness’ score (ρ = 0.612; P < 0.05). Nevertheless, floor ‘dirtiness’ was not retained in the model, which may indicate that soiling of the floor probably did not cause higher odds for lameness in this study. None of the floor characteristics measured in this study had a significant effect on development of lameness within the first 3–5 days of group housing. In contrast, Cador et al. (2014) observed an association between lameness and floor ‘dirtiness’. The lower number of herds in the present study may be one reason why no such association could be found. The general good quality of flooring for all herds probably was the reason why no association between quality and development of lameness could be detected in this study. Slipperiness of the floor differed greatly between herds. Penny et al. (1965) previously measured slipperiness of sow pen floors quantitatively in commercial herds. In the present study, three herds had a slip resistance of less than the recommended minimum of 0.63 (Thorup et al., 2007) whereas five herds had a slip resistance that exceeded the advised maximum of 0.84 (Penny et al., 1965). However, the significance of slipperiness may have been undetectable due to the fact that only the short term effect of the flooring was evaluated.

Similar to the results of Turner et al. (2006), a wide phenotypic variability in skin lesions was found, with disproportionate numbers of skin lesions being present on a minority of the sows. The number of skin lesions is associated with the number of aggressive interactions following the grouping of sows (Barnett et al., 1992). In the present study, aggressive encounters between sows were not identified as a risk factor for development of lameness within the first 3–5 days of group housing. When using lesions as an indicator of post-mixing aggression, only injurious aggressiveness is quantified. However, aggression that does not lead to physical injuries, such as pushing, might also contribute to movement disorders. It would therefore be interesting to conduct observations of (aggressive) behaviour in future lameness research. One advantage of observational studies is that the animals are observed in their natural setting, which results in a high external validity. The disadvantage is that many factors may differ between sows, and between herds, and such studies may have limited the power to detect significant differences. Although the focus of the present study was on housing conditions and post-grouping aggression, also other factors, such as rearing management practices and measures to reduce post-mixing aggression, that were not included in the study, may have been of influence. Observational studies including more herds or experimental studies are necessary to reveal other risk factors for development of lameness and to identify the specific cause of the association between development of lameness and herd size found in this study. Conclusions The estimated incidence of lameness in gestating sows within the first 3–5 days of group housing was 13.1%. Sow dirtiness is a possible proxy indicator of lameness. Increasing the floor space allowance, along with other as yet unknown features of large herds, may diminish development of lameness post-grouping. However, development of lameness did not appear to be influenced by floor characteristics or injurious encounters amongst sows. To prevent sow lameness shortly after introduction into the group, the association of herd size and sow ‘dirtiness’ with development of lameness needs to be explored further. Further research should reveal whether the European Union recommendations on floor space allowance are sufficient to control lameness in group housed sows. Both the role of rearing management practices and measures to reduce aggression should be clarified. Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper. Acknowledgements This study was financed by a grant from ‘The Institute for Innovation through Science and Technology Flanders, Belgium’ (SB091420). The authors are grateful to the participating farmers for their willingness to cooperate in this study. References Andersen, I.L., Bøe, K.E., 1999. Straw bedding or concrete floor for loose-housed pregnant sows: Consequences for aggression, production and physical health. Acta Agriculturae Scandinavica Section A – Animal Science 49, 190–195. Anil, L., Bhend, K.M.G., Baidoo, S.K., Morrison, R., Deen, J., 2003. Comparison of injuries in sows housed in gestation stalls versus group pens with electronic sow feeders. Journal of the American Veterinary Medical Association 223, 1334–1338.

L.M. Pluym et al. / The Veterinary Journal 220 (2017) 28–33

Anil, L., Anil, S.S., Deen, J., Baidoo, S.K., Wheaton, J.E., 2005. Evaluation of well-being, productivity, and longevity of pregnant sows housed in groups in pens with an electronic sow feeder or separately in gestation stalls. American Journal of Veterinary Research 66, 1630–1638. Appleby, M.C., 2005. Welfare challenges in sow housing. Journal of the American Veterinary Medical Association 226, 1334–1336. Arey, D.S., Edwards, S.A., 1998. Factors influencing aggression between sows after mixing and the consequences for welfare and production. Livestock Production Science 56, 61–70. Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Newman, E.A., McCallum, T.H., Chilton, D., 1992. Effects of pen size, partial stalls and method of feeding on welfarerelated behavioural and physiological responses of group-housed pigs. Applied Animal Behaviour Science 34, 207–220. Bonde, M., Rousing, T.J., Badsberg, J.H., Sørensen, J.T., 2004. Associations between lying-down behaviour problems and body condition, limb disorders and skin lesions of lactating sows housed in farrowing crates in commercial sow herds. Livestock Production Science 87, 179–187. Cador, C., Pol, F., Hamoniaux, M., Dorenlor, V., Eveno, E., Guyomarc’h, C., Rose, N., 2014. Risk factors associated with leg disorders of gestating sows in different group-housing systems: A cross-sectional study in 108 farrow-to-finish farms in France. Preventive Veterinary Medicine 116, 102–110. Calderón Díaz, J.C., Fahey, A.G., KilBride, A.L., Green, L.E., Boyle, L.A., 2013. Longitudinal study of the effect of rubber slat mats on locomotory ability, body, limb and claw lesions, and dirtiness of group housed sows. Journal of Animal Science 91, 3940–3954. Chapinal, N., Ruiz de la Torre, J.L., Cerisuelo, A., Gasa, J., Baucells, M.D., Coma, J., Vidal, A., Manteca, X., 2010. Evaluation of welfare and productivity in pregnant sows kept in stalls or in 2 different group-housing systems. Journal of Veterinary Behavior 5, 82–93. Dohoo, I., Martin, W., Stryhn, H., 2009. Veterinary Epidemiologic Research, 2nd ed. VER, Charlottetown, Prince Edward Island, Canada, pp. 395–426. Estienne, M.J., Harper, A.F., Knight, J.W., 2006. Reproductive traits in gilts housed individually or in groups during the first thirty days of gestation. Journal of Swine Health and Production 14, 241–246. Galindo, F., Broom, D.M., 2002. The effects of lameness on social and individual behavior of dairy cows. Journal of Applied Animal Welfare Science 5, 193–201. Geverink, N., Tuyttens, F.A.M., Geenen, H., Geers, R., 2009. Groepshuisvesting voor Zeugen en Gelten: Repercussie op Algemeen Welzijn, Gezondheid en Milieu, 1st ed. Federale Overheidsdienst Volksgezondheid, Veiligheid van de Voedselketen en Leefmilieu, Brussels, Belgium, p. 171. Gjein, H., Larssen, R.B., 1995a. Housing of pregnant sows in loose and confined systems – A field-study. 1. Vulva and body lesions, culling reasons and production results. Acta Veterinaria Scandinavica 36, 185–200. Gjein, H., Larssen, R.B., 1995b. The effect of claw lesions and claw infections on lameness in loose housing of pregnant sows. Acta Veterinaria Scandinavica 36, 451–459. Heinonen, M., Oravainen, J., Orro, T., Seppä-Lassila, L., Ala-Kurikka, E., Virolainen, J., Tast, A., Peltoniemi, O.A.T., 2006. Lameness and fertility of sows and gilts in randomly selected loose-housed herds in Finland. Veterinary Record 159, 383–387. Hosmer, D.W., Lemeshow, S., 2000. Applied Logistic Regression, 2nd ed. John Wiley & Sons, New York, pp. 92–128.

33

KilBride, A., Gillman, C., Green, L., 2009. A cross-sectional study of the prevalence of lameness in finishing pigs, gilts and pregnant sows and associations with limb lesions and floor types on commercial farms in England. Animal Welfare 18, 215–224. Knox, R., Salak-Johnson, J., Hopgood, M., Greiner, L., Connor, J., 2014. Effect of day of mixing gestating sows on measures of reproductive performance and animal welfare. Journal of Animal Science 92, 1698–1707. Kroneman, A., Vellenga, L., Van der Wilt, F.J., Vermeer, H.M., 1993. Field research on veterinary problems in group-housed sows – A survey of lameness. Journal of Veterinary Medicine Series A 40, 704–712. Madec, F., Cariolet, R., Dantzer, R., 1986. Relevance of some behavioral criteria concerning the sow (motor-activity and water-intake) in intensive pig farming and veterinary practice. Annales de Recherches Veterinaires 17, 177–184. Main, D.C.J., Clegg, J., Spatz, A., Green, L.E., 2000. Repeatability of a lameness scoring system for finishing pigs. Veterinary Record 147, 574–576. McKee, C.I., Dumelow, J., 1995. A review of the factors involved in developing effective non-slip floors for pigs. Journal of Agricultural Engineering Research 60, 35– 42. Penny, R.H., Osborne, A.D., Wright, A.I., Stephens, T.K., 1965. Foot-rot in pigs: Observations on the clinical disease. Veterinary Record 77, 1101–1108. Pluym, L.M., Van Nuffel, A., Dewulf, J., Cools, A., Vangroenweghe, F., Van Hoorebeke, S., Maes, D., 2011. Prevalence and risk factors of claw lesions and lameness in pregnant sows in two types of group-housing. Veterinarni Medicina 56, 101– 109. Pluym, L.M., Van Nuffel, A., Van Weyenberg, S., Maes, D., 2013. Prevalence of lameness and claw lesions during different stages in the reproductive cycle of sows and the impact on reproduction results. Animal: An International Journal of Animal Bioscience 7, 1174–1181. Salak-Johnson, J.L., Niekamp, S.R., Rodriguez-Zas, S.L., Ellis, M., Curtis, S.E., 2007. Space allowance for dry, pregnant sows in pens: Body condition, skin lesions, and performance. Journal of Animal Science 85, 1758–1769. Street, B.R., Gonyou, H.W., 2008. Effects of housing finishing pigs in two group sizes and at two floor space allocations on production, health, behavior, and physiological variables. Journal of Animal Science 86, 982–991. Thorup, V.M., Tøgersen, F.A., Jørgensen, B., Jensen, B.R., 2007. Biomechanical gait analysis of pigs walking on solid concrete floor. Animal: An International Journal of Animal Bioscience 1, 708–715. Turner, S.P., Farnworth, M.J., White, I., Brotherstone, S., Mendl, M., Knap, P., Penny, P., Lawrence, A.B., 2006. The accumulation of skin lesions and their use as a predictor of individual aggressiveness in pigs. Applied Animal Behaviour Science 96, 245–259. Webb, N.G., 1984. Compressive stresses on, and the strength of the inner and outer digits of pigs’ feet, and the implications for injury and floor design. Journal of Agricultural Engineering Research 30, 71–80. Webb, N.G., Nilsson, C., 1983. Flooring and injury −An overview. In: Baxter, S.H., Baxter, M.R., McCormack, J.A.C. (Eds.), Farm Animal Housing and Welfare. Kluwer Academic Publishers Group, Dordrecht, The Netherlands, pp. 129–136. Willgert, K.J.E., Brewster, V., Wright, A.J., Nevel, A., 2014. Risk factors of lameness in sows in England. Preventive Veterinary Medicine 113, 268–272. Zurbrigg, K., Blackwell, T., 2006. Injuries, lameness and cleanliness of sows in four group-housing gestation facilities in Ontario. Journal of Swine Health and Production 14, 202–206.