A cross-sectional study of lactating sows in 71 commercial pig herds in England was carried out to estimate the prevalence of foot lesions and abnormal posture and investigate the associated risks. Observational data were collected on the standing posture, lesions on the hind left foot, and the pen the sow was housed in for 233 sows. The prevalence of foot lesions was 76.8% and 10.8% of lactating sows had abnormal posture. Sows housed on slatted floors during pregnancy had an associated increased risk of heel flaps but a reduced risk of toe erosion, compared with sows housed on solid concrete floors with bedding during pregnancy. There was an increased risk of abnormal posture in lactating sows housed on slatted floors during pregnancy compared with those housed on solid concrete floors with straw bedding during pregnancy. In addition, there was an increased risk of abnormal posture in lactating sows with wounds on the limbs and erosions on the toe and the heel/sole. The positive association between abnormal posture and foot lesions suggests that these lesions might be painful. Housing sows on solid floors during pregnancy might reduce the risk of abnormal posture, and certain foot lesions, in sows during lactation. However, posture is a less sensitive measure of lameness than gait, and the nature of the causal relationship between foot lesions and lameness requires further investigation.

Footrot is a costly endemic disease of sheep. This study investigates the potential to decrease its prevalence through selective breeding for decreased lesion score. Pedigreed mule and Scottish Blackface (SBF) ewes were scored for lesions on each hoof on a 0 to 4 scale for up to 2 (SBF ewes) or 4 (mules) times over 2 years. One score was obtained for SBF lambs. An animal was deemed to have lesions (severe lesions) if at least one hoof had a score of at least 1 (2). The prevalence of lesions was 34% in lambs, 17% in SBF ewes and 51% in mules. The heritability of lesions (severe lesions) analysed as repeated measurements of the same trait in a threshold model was 0.19 (0.26) in SBF ewes and 0.12 (0.19) in mules. Estimates for the sum and maximum of scores as well as the number of feet affected were much lower, as were estimates for permanent animal effects (i.e. non-genetic effects associated with an animal). When successive scores on the same animal were analysed as correlated traits, heritability estimates for most traits tended to be higher, except for severe footrot in mules where estimates varied greatly over time. The phenotypic correlations between successive scores in SBF ewes were close to 0, genetic correlations were moderately positive (0.18 to 0.55). Correlations in mules were generally of a similar size, but some genetic correlations were higher (up to 0.92). There was a clear trend for heritabilities for lesions and severe lesions to increase with higher prevalence of lesions, even when analysed in a threshold model. Heritability estimates for traits that combine scores over several events in mules, identifying the more persistently affected animals, ranged from 0.12 to 0.23 with the highest estimates for the average number of feet that were (severely) affected in animals scored for a minimum at two events. The heritability of all lesion traits in lambs was estimated as 0. It is concluded that selection for lower lesions is possible in ewes but not lambs, and that a simple binary score at an animal level is at least as effective as a comprehensive score at hoof level. Given the low repeatability of lesion scores, repeated measures over time will improve effectiveness of selection. Selection across environments (flocks, seasons) with different prevalences of lesions scores will need to take account of variation in the heritability.