Profile and genetic parameters of dairy cattle locomotion score and lameness across lactation

Animal (2014), 8:1, pp 20 27 The Animal Consortium 2013 doi:10.1017/s1751731113001717 animal Profile and genetic parameters of dairy cattle locomotion score and lameness across lactation A. Kougioumtzis 1, G. E. Valergakis 1, G. Oikonomou 1,2, G. Arsenos 1 and G. Banos 1,3 1 Laboratory of Animal Husbandry, Faculty of Veterinary Medicine, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece; 2 College of Veterinary Medicine, Cornell University, Ithaca, NY, USA; 3 Scotland s Rural College/Roslin Institute, Easter Bush, Midlothian EH25 9RG, Scotland, UK (Received 21 January 2013; Accepted 02 August 2013; First published online 23 October 2013) This study investigated the profile of locomotion score and lameness before the first calving and throughout the first (n = 237) and second (n = 66) lactation of 303 Holstein cows raised on a commercial farm. Weekly heritability estimates of locomotion score and lameness, and their genetic and phenotypic correlations with milk yield, body condition score, BW and reproduction traits were derived. Daughter future locomotion score and lameness predictions from their sires breeding values for conformation traits were also calculated. First-lactation cows were monitored weekly from 6 weeks before calving to the end of lactation. Second-lactation cows were monitored weekly throughout lactation. Cows were locomotion scored on a scale from one (sound) to five (severely lame); a score greater than or equal to two defined presence of lameness. Cows weekly body condition score and BW was also recorded. These records were matched to corresponding milk yield records, where the latter were 7-day averages on the week of inspection. The total number of repeated records amounted to 12 221. Data were also matched to the farm s reproduction database, from which five traits were derived. Statistical analyses were based on uni- and bivariate random regression models. The profile analysis showed that locomotion and lameness problems in first lactation were fewer before and immediately after calving, and increased as lactation progressed. The profile of the two traits remained relatively constant across the second lactation. Highest heritability estimates were observed in the weeks before first calving (0.66 for locomotion score and 0.54 for lameness). Statistically significant genetic correlations were found for first lactation weekly locomotion score and lameness with body condition score, ranging from 0.31 to 0.65 and from 0.44 to 0.76, respectively, suggesting that cows genetically pre-disposed for high body condition score have fewer locomotion and lameness issues. Negative (favourable) phenotypic correlations between first lactation weekly locomotion score/lameness and milk yield averaged 0.27 and 0.17, respectively, and were attributed to management factors. Also a phenotypic correlation between lameness and conception rate of 0.19 indicated that lame cows were associated with lower success at conceiving. First-lactation daughter locomotion score and/or lameness predictions from sires estimated breeding values for conformation traits revealed a significant linear effect of rear leg side view, rear leg rear view, overall conformation, body condition score and locomotion, and a quadratic effect of foot angle. Keywords: locomotion score, lameness, dairy cattle, lactation, genetic parameters Implications Results from this study reveal insights into the profile of cow locomotion score and lameness throughout lactation. Both traits are of considerable interest and importance to dairy farmers and breeders around the world. The outcomes of the study also indicate the existence of heritable variation in the two traits which can be used to genetically improve locomotion and lameness without adversely affecting other E-mail: geval@vet.auth.gr important traits. Practical implementation of the results can contribute to better on-farm management and enhanced genetic selection and breeding programmes. Introduction Lameness is a major health and welfare issue for dairy cattle worldwide (Greenough, 2007). Together with reproductive failure and mastitis, lameness is one of the most common causes of cow culling (United States Department of Agriculture, 2007). Reported incidence and prevalence range from 20

Profile and genetic parameters of dairy cattle lameness 23.7% to 54.6% and 15.0% to 62.0%, respectively (Whitaker et al., 2000; Haskell et al., 2006; Archer et al., 2010). In the Netherlands, the cost per case has been estimated to range from 48 to 886, including treatment associated expenses, milk yield loss and discarded milk, increased calving intervals and culling (Noordhuizen, 2012). Lameness is a symptom, not a disease. Regardless of the actual cause, it is perceived as impaired locomotion of variable severity (Greenough, 2007). In order to quantify this impairment, various 3- to 6-point scales have been developed by different research groups, based mainly on gait and posture characteristics (Amory et al., 2006; Archer et al., 2010). These are generally considered credible descriptors of lameness severity; when one such system was tested, it was found to be well correlated with foot lesions (Bicalho et al., 2007a). Locomotion scoring is a widespread management practice, providing useful epidemiological data, both for monitoring and confronting the problem (Haskell et al., 2006). However, a pattern of lameness dynamics before first calving and during a complete lactation is lacking as published research has mainly focused on the first stage of lactation, trying: (a) to identify correlations with milk production and reproductive performance or (b) to ascertain the influence of putative risk factors. Causes of lameness are classified as either contagious or as a disruption of claw horn formation and are actually multifactorial in nature (Greenough, 2007). Several management-related issues like farming system, nutrition, cow comfort and hygienic conditions are well-recognised risk factors and they are specifically targeted during farm investigations dealing with increased prevalence of lameness (Cook, 2003; Amory et al., 2006; Espejo and Endres, 2007). Unambiguously, lameness is a heritable trait, as well. Heritability estimates of locomotion scores either for diagnosing lameness (Van Dorp et al., 2004) or as a linear trait assessed as part of a linear type classification scheme (Onyiro and Brotherstone, 2008; Laursen et al., 2009; van der Linde et al., 2010) range from 0.05 to 0.14. Heritability of lameness itself has been estimated to range from 0.07 to 0.10 (Boettcher et al., 1998; Pryce et al., 1999; Zwald et al., 2004). Early heritability estimates of specific foot diseases/lesions (Koenig et al., 2005; van der Waaij et al., 2005) ranged from 0.01 to 0.11; more recent ones range from 0.03 to 0.07 (Onyiro et al., 2008; van der Linde et al., 2010; Buch et al., 2011). Reported genetic correlations with body condition score (Van Dorp et al., 2004; Onyiro et al., 2008) and reproductive performance (Buch et al., 2011; Zink et al., 2011) were unfavourable; on the other hand, they were favourable with milk yield and milk component traits (Van Dorp et al., 1998; Buch et al., 2011). None of these studies discriminate between before and after first calving or between different stages of lactation. Calving and lactation onset is a stressful event and environmental effects are expected to be greater around and, cumulatively, after lactation peak. Most of the research papers cited above also report significant genetic correlations of lameness, locomotion scores and specific diseases/lesions with feet conformation traits (feet and leg composite, rear leg side view, rear leg rear view and foot angle) based on cow records. Swalve et al. (2008) derived prediction equations of daughter locomotion score and lameness from their sires estimated breeding values for conformation traits. This could facilitate farmers decisions regarding sire selection, as part of a multi-trait selection index, in order to decrease the incidence and prevalence of lameness in their herds. The objective of this study was: (i) to determine the profile of locomotion score and lameness before the first calving and throughout the first and second lactation, (ii) to investigate whether heritability estimates and their correlation with milk yield, body condition score, BW and reproduction of these traits vary throughout the same period, and (iii) to derive cow locomotion and lameness predictions from their sires breeding values for conformation traits. Material and methods Animals and data Data were derived from a large commercial Holstein herd in Northern Greece; the farm was closely monitored by the authors research team from January 2008 until April 2010. Cows were housed in two similar free-stall barns; the lying surfaces were covered with rubber mattresses without extra bedding. Walking surface was grooved concrete and no outside exercise area was provided. Cows were fed, twice a day, a total mixed ration to meet their energy and protein requirements. Ration formulation was based on US National Research Council recommendations (NRC, 2001). Cows were bred using artificial insemination, either after a natural oestrus or after being synchronized using the Ovsynch protocol, described by (Stevenson et al., 1996). A total of 303 cows (237 in their first and 66 in their second lactation) were included in the study. These cows had been imported from five different countries (Denmark, France, Germany, the Netherlands and Switzerland) as pregnant heifers and were daughters of 155 proven sires. About 10% of the sires were common across the five countries. All cows had complete pedigree. All cows were monitored on a weekly basis throughout their respective lactation leading to the collection of a total of 12 221 repeated records on various traits as described below. The locomotion of each cow was evaluated weekly throughout the entire lactation by the leading author, a trained veterinarian, during the animal s exit from the milking parlour, using the method described by Rajkondawar et al. (2006). Briefly, cows received a locomotion score from one to five (in increments of one), where one denoted an animal with a sound, uninhibited and symmetrical gait, and five a seriously lame animal, reluctant or incapable of moving or standing up. For first-lactation cows, weekly locomotion evaluation started 6 weeks before their first calving. Cows with locomotion score equal to one were considered healthy and lameness-free, whereas cows with a locomotion score 2 were considered lame. This is a conservative definition 21

Kougioumtzis, Valergakis, Oikonomou, Arsenos and Banos of the condition that combines mild and serious cases of lameness into a single category. All cows diagnosed as lame were hoof trimmed by the leading author; lesions were recorded and treated as necessary. Body condition score was recorded immediately after the locomotion evaluation by the leading author using the 1- to 5-point scale of Ferguson et al. (1994) in increments of 0.25. The ends of this scale refer to emaciated (1) and obese (5) animals, respectively. An estimation of BW was also recorded at the time of body condition evaluation as a weight scale was not available on the farm. BW was assessed with the indirect method proposed by Heinrichs et al. (1992) based on the measurement of the heart girth. The following equation was used to assess BW: BW ¼ 102:71 ð2:876 HGÞ + ð0:02655 HG 2 Þ where BW is the body weight (kg); HG the heart girth (cm). Daily milk yield records were available in the database of the farm (DeLaval in-line milk metres, serviced twice a year). A 7-day average was calculated for each week of locomotion score, lameness, body condition score and BW record. The farm database was also used to derive the following reproductive traits: calving interval (between first and second calving for first-lactation cows and between second and third calving for second-lactation cows), conception rate at first insemination and in the first 305 days of lactation, number of inseminations per conception, and interval from calving to conception. Finally, estimated breeding values of the sires of these cows were obtained from Interbull (International Bull Evaluation Service, 2012) for the following traits: rear leg side view, rear leg rear view, foot angle, overall body conformation, overall feet and leg conformation, body condition score, and locomotion. Since sires were from different countries, their estimated breeding values were expressed on the US scale and base for comparability and consistency. Statistical analyses Repeated records of locomotion score and lameness were analysed with the following model; each trait was analysed separately: Y ijkmnopq ¼ YS i + C j + L o + M p + S q + a 1 age + X2 n¼0 + X1 n¼0 b n P n W m + X2 n¼0 A kn P n W m PE kn P n W m + e ijkmnpq ð1þ where Y ijkmnopq is the locomotion score or lameness record of cow k in week from calving m; YS i the fixed effect of yearseason of calving i (four levels); C j the fixed effect of country of origin j of cow (five levels); L o the fixed effect of number of lactation o (two levels); M p the fixed effect of calendar month of the year p (12 levels); S q the fixed effect of barn q (two levels); a 1 the linear regression coefficients on age at calving (age); W m the week m from calving; b n the fixed regression coefficient on week from calving; A kn the random regression coefficient on week from calving associated with the additive genetic effect of cow k (including all pedigree amounting to 2139 animals in total); P n the nth orthogonal polynomial of week from calving (n the order of polynomial); PE kn the random regression coefficient on week from calving associated with the permanent environment effect of cow k; e ijkmnopq the random residual term. In model 1, the fixed regression coefficient was associated with an overall average lactation curve per lactation for locomotion score and lameness, whereas the random regression was associated with each individual cow s deviation from the overall curve. The fixed effects in the model were fitted after preliminary analyses showed they had a statistically significant effect (P < 0.05) on the traits. The final order of random polynomial (third for either trait) was determined with the use of the log-likelihood test in sequential analyses of gradually increasing orders. Polynomial order is 3 for animal (0, 1, 2) and 2 for permanent environment (0, 1). Four measurement error classes were defined using the time relative to calving as: all weeks before the first calving and weeks 1 to 2, 3 to 39 and 40 to 45 after calving. The definition of these classes aimed to capture the differences in various periods associated with pre-partum, immediate post-partum, main lactation and late lactation. The specific threshold, albeit somewhat subjective, coincided with considerable changes in the variance of locomotion score. Different residual variances were estimated for each measurement error class, while co-variances between these classes were assumed to be zero. Aprobitfunctionwasfitted to model 1 for the analysis of lameness to account of the binary nature of the trait. In this case, results were expressed on the underlying liability scale. Estimates of (co)variance components from model 1 were used to calculate heritabilities for each trait and week from calving. The relevant magnitude of these results was confirmed in a series of repeatability model analyses of data spanning the same time as each of the four measurement error classes. A series of bivariate analyses using model 1 were also conducted to calculate genetic and phenotypic correlations of locomotion score and lameness on the one hand with milk yield, body condition score and BW on the other. Probit function for the analysis of lameness to account of the binary nature of the trait was used in these analyses, too. In another series of bivariate analyses, genetic and phenotypic correlations of locomotion score and lameness with reproductive traits were estimated. Locomotion score and lameness repeated records were analysed with model 1 whilst reproduction was analysed with the following model: Y ijknopq ¼ YS i + C j + L o + S q + M p + G n + a 1 age + a r dm + A k + e ijknopq ð2þ Y ijknopq is the reproductive trait of cow k; YS i the fixed effect of year-season of calving i (four levels); C j the fixed effect of country of origin j of cow (five levels); L o the fixed effect of number of lactation o (two levels); S q the fixed effect of barn q (two levels); M p the fixed effect of calendar month p of the first artificial insemination of the cow (12 levels); 22

Profile and genetic parameters of dairy cattle lameness Table 1 Average locomotion score (1 to 5 scale) and lameness incidence (0 to 1 scale) of 237 first-lactation and 66 second-lactation dairy cows observed in weekly intervals, in various lactation stages Locomotion score Lameness incidence Stage Lactation 1 Lactation 2 Lactation 1 Lactation 2 Before calving 0.97 0.11 1 to 2 weeks of post calving 1.19 1.58 0.24 0.47 3 to 39 weeks of post calving 1.60 1.60 0.50 0.46 40 to 45 weeks of post calving 1.69 1.57 0.56 0.43 G n the fixed effect indicating whether the cow participated in an oestrus synchronisation scheme implemented on the farm (two level); a 1 the linear regression coefficients on age at calving (age); a r the linear regression coefficients on days in milk when the first artificial insemination of the cow took place (dm); A k the random genetic effect of cow k (including all pedigree); e ijknopq the random residual term. The impact of the sires estimated breeding values for conformation traits on daughter first-lactation locomotion score and lameness was assessed by adding a linear regression on the former in model 1. A quadratic regression was also fitted for traits with an intermediate optimum, namely rear leg side view and foot angle. All analyses were conducted using the statistical software package ASREML (Gilmour et al., 2006). Figure 1 Locomotion score profile of 237 first-lactation and 66 secondlactation dairy cows across lactation; random regression mixed model solutions fitting third order polynomials. Results and discussion Adjusted average locomotion score and lameness incidence in various time periods are shown in Table 1. The periods coincide with the measurement error classes defined in model 1 that aimed to capture variance differences in distinct periods associated with calving and the ensuing lactation. Both locomotion score and lameness appear to be less of a problem before or immediately after the first calving compared with the main duration of first lactation and the whole second lactation. A likely reason is that high milk production demands in the latter stages of lactation influence the cow s metabolic profile leading to increased susceptibility to these conditions. Boettcher et al. (1998) reported higher lameness prevalence during the earliest stages of lactation but comparisons with our study are difficult due to different demarcation of lactation stages. On the other hand, our results are in agreement with those of Capion et al. (2008) who report lameness prevalence to be lowest before parturition with a gradual increase through the latter stages of lactation. The trends observed in Table 1 are confirmed in Figures 1 and 2 that illustrate the fixed curves for locomotion score and lameness, respectively, derived from the analyses with model 1. These curves are adjusted for all sources of systematic variation that were included in the model as well as the correlation between successive weekly assessments of locomotion score and lameness. Thus, the effect on the locomotion score of a cow in a given week on the score of the following week is effectively accounted for by the function Figure 2 Lameness profile across lactation of 237 first-lactation and 66 second-lactation dairy cows; random regression mixed model solutions fitting third order polynomial. (polynomial) used in model 1 to describe the trajectory of the trait across time. To our knowledge, this is the first report of adjusted locomotion score and lameness profile across a cow s lactation that also includes data before calving. Weekly estimates of the genetic variance for locomotion score and lameness are shown in Figures 3 and 4, respectively. These estimates were derived from the random regression model 1 and were statistically greater than zero (P < 0.05) in all cases, demonstrating the presence of genetic factors that control the animal s capacity for uninhibited movement and resistance to lameness. The estimates before first calving are particularly noteworthy, since no such results have been reported in the literature. Although relatively few locomotion and lameness problems were associated with this period (Figures 1 and 2), the amount of genetic variance 23

Kougioumtzis, Valergakis, Oikonomou, Arsenos and Banos present suggests that selection would be particularly effective then. Estimates of first-lactation heritability derived from the genetic variances discussed above are shown in Table 2. These are average estimates of weekly heritabilities in the designated time periods and were all significantly greater than zero (P < 0.05). Average weekly heritability estimates for the entire period for locomotion score and lameness were 0.39 ± 0.06 and 0.26 ± 0.06. The genetic variance estimates (Figures 3 and 4) were consistent with the heritability of locomotion score and lameness and were greatest before and immediately after the first calving. Heritability estimates for milk yield, body condition score and BW are also shown for comparison. Sizeable heritability estimates for locomotion score and lameness derived throughout lactation suggest that both traits are amenable to improvement with genetic selection, nearly as much as the other traits shown in Table 2, which Figure 3 Estimates of the genetic variance of locomotion score of 237 first-lactation and 66 second-lactation dairy cows by week from calving; standard errors ranged from 0.03 to 0.12. Figure 4 Estimates of the genetic variance of lameness of 237 firstlactation and 66 second-lactation dairy cows by week from calving; standard errors ranged from 0.01 to 0.04. already feature in the breeding programmes in many countries. Similar to the conclusion of van der Waaij et al. (2005), first-lactation locomotion score and lameness records are suitable indicators for future claw health. Weekly heritability estimates across the second lactation (not shown) followed closely the trend in genetic variance (Figures 3 and 4). Average heritability across the second lactation was 0.29 ± 0.06 for locomotion score and 0.18 ± 0.05 for lameness. As this is a one-herd study in a controlled environment, residual variances are expected to be low allowing the genetic variance estimates to account a greater proportion of the total phenotypic variance. Therefore, locomotion score and lameness heritability estimates derived in the present study are higher compared with literature estimates based on field studies with many herds and more diverse environments (0.05 to 0.14), (Van Dorp et al., 2004; van der Linde et al., 2010). Another novelty of this study is the longitudinal analysis of repeated locomotion score and lameness records per cow throughout lactation using random regression models. No previous estimates of such analysis for these traits were found in the literature. Genetic and phenotypic correlation estimates between locomotion score and lameness on the one hand, and body condition score on the other are shown in Figure 5. These are weekly estimates derived from bivariate analyses based on model 1 and pertain to the period before first calving and throughout first lactation. Genetic correlations were statistically significant (P < 0.05) beyond week 12 and 8 of lactation for locomotion score and lameness, respectively, while phenotypic correlations were statistically significant beyond week 4 for both traits. Negative genetic correlations of locomotion score and lameness with body condition score (shown in Figure 5), suggest that high body condition scores are associated with healthy cow movement and reduced lameness on any week of lactation. At the genetic level, this indicates that cows pre-disposed to maintain good levels of body condition score are also less susceptible to locomotion and lameness problems. This is consistent with the results of Boettcher et al. (1998) and Van Dorp et al. (2004) in studies conducted with Holstein cows in United States and Canada. Statistically significant (P < 0.05) negative genetic correlations were also found between body condition score on 1 week of lactation and locomotion score and lameness on the following week. These phenotypic and genetic correlation estimates varied from 0.32 to 0.64 for locomotion score and 0.46 to 0.75 for lameness. Table 2 Average heritability estimates of five traits from 237 first-lactation dairy cows in various stages of lactation (standard errors in parentheses) Stage Locomotion Lameness Milk yield BCS BW Before calving 0.66 (0.07) 0.54 (0.08) 0.59 (0.07) 0.45 (0.08) 1 to 2 weeks of post calving 0.58 (0.07) 0.37 (0.07) 0.32 (0.07) 0.58 (0.07) 0.45 (0.08) 3 to 39 weeks of post calving 0.34 (0.06) 0.25 (0.06) 0.50 (0.08) 0.58 (0.08) 0.36 (0.09) 40 to 45 weeks of post calving 0.44 (0.09) 0.29 (0.09) 0.56 (0.08) 0.64 (0.07) 0.39 (0.10) BCS = body condition score. 24

Profile and genetic parameters of dairy cattle lameness Phenotypic correlations of milk yield with locomotion score and lameness are shown in Figure 6. These results pertain to phenotypic correlation estimates on the same week in lactation and are all statistically significant (P < 0.05). Genetic correlations between these traits (not shown) were not statistically different from zero (P > 0.05). A positive genetic correlation between milk yield and lameness is reported in the literature (Van Dorp et al., 1998; 0.0-0.1-0.2-0.3-0.4-0.5-0.6-0.7-0.8-0.9-1.0 Week from first calving -6-3 1 4 7 101316192225283134374043 gen_lc gen_lm phen_lc phen_lm Figure 5 Genetic and phenotypic correlations between locomotion score and body condition score (gen_lc, phen_lc) and between lameness and body condition score (gen_lm, phen_lm) by week from first calving (237 cows); statistically significant (P < 0.05) estimates for weeks 12 or greater (genetic-locomotion), 8 or greater (genetic-lameness) and 4 or greater (phenotypic-both traits); standard errors ranged from 0.05 to 0.22. Figure 6 Phenotypic correlations of milk yield with locomotion score and lameness by week of first-lactation (237 cows); all estimates were statistically significant (P < 0.05); standard errors ranged from 0.03 to 0.07. Koenig et al., 2005; Buch et al., 2011). Our results suggest that phenotypic correlations in Figure 6 mainly illustrate an unfavourable environmental correlation; lame cows have lower feed intake and consequently, lower milk yield. This observation alludes to proper management practices favouring all traits in question. Genetic and phenotypic correlations of body condition score and milk yield with locomotion score and lameness in the second lactation were not statistically different from zero (P > 0.05). The same was true for correlations of BW with locomotion score and lameness in both lactations. Genetic and phenotypic correlations of locomotion score and lameness with reproductive traits were not statistically different from zero (P > 0.05). The only exception was an average phenotypic correlation between lameness and conception rate in the first 305 days of lactation of 0.19 ± 0.07, indicating that lame cows are associated with lower success at conceiving. Zink et al. (2011) also found a negative genetic correlation ( 0.34) between locomotion scores and days from calving to conception, both for first and second parity cows. This result was consistent with Pryce et al. (2000) who found that excellent locomotion was correlated with shorter calving intervals. Bicalho et al. (2007b) also concluded that lame cows have poor reproductive performance and increased odds to be culled. Significant heritable variation found for locomotion score and lameness in the present study combined with the genetic and phenotypic correlation results discussed above warrant genetic selection for the improvement of the two traits that is not expected to adversely affect other important animal functions. Prediction equations of first-lactation daughter locomotion score and lameness from the sires estimated breeding values for conformation are shown in Table 3. Since the cows considered in the present study had not been included in the estimation of their sires breeding value, the regressions derived represent an authentic association of the two variables based on independent datasets. Of interest are the quadratic relationships with traits exhibiting intermediate optima, namely scores from 4 to 6 on the 1 to 9 linear scale system (Holstein Association USA, 2012). Optimum (intermediate) foot angle conformation is indeed related to improved locomotion and less lameness; steeper Table 3 Regression coefficients of 237 first-lactation cow locomotion score and lameness on their sires (155 bulls) estimated breeding values for conformation traits Locomotion score Lameness Trait Intercept Linear Quadratic Intercept Linear Quadratic Rear leg side view 0.68* 0.0251* 0.0021 0.12 0.0143* 0.0056 Rear leg rear view 0.68* 0.0057 0.13 0.0107* Foot angle 0.65* 0.0031 0.0193* 0.15 0.0113 0.0150* Overall conformation 0.64* 0.0383* 0.15 0.0206 Overall feet and leg 0.67* 0.0117 0.14 0.0088 Locomotion 0.67* 0.0580* 0.14 0.0328* Body condition score 0.75* 0.0473* 0.09 0.0273* *Statistically different from zero (P < 0.05). 25

Kougioumtzis, Valergakis, Oikonomou, Arsenos and Banos angles are preferable than lower ones. On the other hand, the quadratic relationships of rear leg side view, which is also an intermediate optimum trait, were not statistically significant from zero (P > 0.05); instead, the linear ones were, suggesting that straight rear legs are associated with improved locomotion and less lameness. Van Dorp et al. (1998 and 2004) reached the same conclusion. Furthermore, it has been suggested that such conformation is also associated with a longer productive life (Holstein Association USA, 2012). Lame cows have increased odds of being culled (Bicalho et al., 2007b). Favourable rear leg, rear view conformation was significantly associated (P < 0.05) with lowerincidenceratesoflamenessinthisstudy;thiswasalso the case(p < 0.05) with overall conformation regarding locomotion score. Feet and leg scores showed a favourable but statistically not significant (P > 0.05) linear relationship with locomotion score and lameness in this study. Selection for improved conformation of feet and legs is considered to result in cows less vulnerable to lameness (Koenig et al., 2005; Onyiro et al., 2008; van der Linde et al., 2010). Many authors also concluded that inclusion of specific claw health traits in a lameness index would result in faster genetic progress (Koenig et al., 2005; van der Linde et al., 2010; Buch et al., 2011). Swalve et al. (2008), working with sires estimated breeding values, concluded that feet and legs conformation traits would not be of great aid in reducing lameness. In the US system, the feet and legs composite index (FLC = 0.5 (0.48 foot angle + 0.37 rear legs rear view 0.15 rear legs side view) + 0.5 (feet and legs score)) may underestimate the effect of rear leg side view scores and this could weaken its role as a lameness predictor. The favourable relationship of locomotion as a conformation trait with lameness was expected; that was the original intention of this trait. Finally, based on our results, the association of higher body condition scores with less lameness was evident once again, consistently with earlier studies (Boettcher et al., 1998; Van Dorp et al., 2004). If a lameness index is to be constructed, it seems that it should include properly weighted feet and leg conformation traits, body condition scores and claw health traits (when available). Until then, farmers and their consultants have to base selection and mating decisions on single or provisionally combined individual traits of known favourable effects. As mentioned earlier, the present study was based on data from a single herd of closely monitored animals. This allowed the opportunity to collect repeated observations on the same cow in a controlled environment for a detailed study of their locomotion score and lameness profile across lactation. The trade-off was that only a limited number of animals were included in the study. Clearly the utility of this study is different from large-scale field studies based on multiple herds and thousands of animals with single or very limited number of lameness records. Results from both types of studies are needed to derive a thorough understanding of the dynamics of these conditions and properly utilise it in genetic selection and improvement schemes. The fact that the heritability of lameness appears to be much higher around first calving that is, when most management issues related with lactation stress have not been raised, yet, could prove a useful tool, in this respect. Acknowledgements The VIVARTIA dairy farm for offering access to their resources and the Interbull Centre for providing sire estimated breeding values for this research study; the first author acknowledges financial support from the Greek State Scholarships Foundation. References Amory JR, Kloosterman P, Barker ZE, Wright JL, Blowey RW and Green LE 2006. Risk factors for reduced locomotion in dairy cattle on nineteen farms in The Netherlands. Journal of Dairy Science 89, 1509 1515. Archer SC, Green MJ and Huxley JN 2010. Association between milk yield and serial locomotion score assessments in UK dairy cows. Journal of Dairy Science 9, 4045 4053. Bicalho RC, Cheong SH, Cramer G and Guard CL 2007a. Association between a visual and an automated locomotion score in lactating Holstein cows. Journal of Dairy Science 90, 3294 3300. Bicalho RC, Vokey F, Erb HN and Guard CL 2007b. Visual locomotion scoring in the first seventy days in milk: impact on pregnancy and survival. Journal of Dairy Science 90, 4586 4591. Boettcher PJ, Dekkers JCM, Warnick LD and Wells SJ 1998. Genetic analysis of clinical lameness in dairy cattle. Journal of Dairy Science 81, 1148 1156. Buch LH, Sorensen AC, Lassen J, Berg P, Eriksson J-A, Jakobsen JH and Sorensen MK 2011. Hygiene-related and feed-related hoof diseases show different patterns of genetic correlations to clinical mastitis and female fertility. Journal of Dairy Science 94, 1540 1551. Capion N, Thamsborg SM and Enevoldsen C 2008. Conformation of hind legs and lameness in Danish Holstein heifers. Journal of Dairy Science 91, 2089 2097. Cook NB 2003. Prevalence of lameness among dairy cattle in Wisconsin as a function of housing type and stall surface. Journal of the American Veterinary Medical Association 223, 1324 1328. Espejo LA and Endres MI 2007. Herd-level risk factors for lameness in highproducing Holstein cows housed in freestall barns. Journal of Dairy Science 90, 306 314. Ferguson JD, Galligan DT and Thomsen N 1994. Principal descriptors of body condition score in Holstein cows. Journal of Dairy Science 77, 2695 2703. Gilmour AR, Gogel BJ, Cullis BR, Welham SJ and Thompson R 2006. ASREML user guide, release 2.0. VSN International Ltd, Hemel Hempstead, UK. Greenough PR 2007. Bovine laminitis and lameness, 1st edition. Elsevier Limited, Philadelphia, PA, USA. Haskell MJ, Rennie LJ, Bowell VA, Bell MJ and Lawrence AB 2006. Housing system, milk production, and zero-grazing effects on lameness and leg injury in dairy cows. Journal of Dairy Science 89, 4259 4266. Heinrichs AJ, Rogers GW and Cooper JB 1992. Predicting bodyweight and wither height in Holstein heifers using body measurements. Journal of Dairy Science 75, 3576 3581. Holstein Association USA 2012. Linear type evaluations. Retrieved 25 November 2012 from www.holsteinusa.com. International Bull Evaluation Service 2012. Retrieved 10 March 2012 from www. interbull.org. Koenig S, Sharifi AR, Wentrot H, Landmann D, Eise M and Simianer H 2005. Genetic parameters of claw and foot disorders estimated with logistic models. Journal of Dairy Science 88, 3316 3325. Laursen MV, Boelling D and Mark T 2009. Genetic parameters for claw and leg health, foot and leg conformation and locomotion in Danish Holsteins. Journal of Dairy Science 92, 1770 1777. Noordhuizen J 2012. Dairy herd health and management. A guide for veterinarians and dairy professionals. Context Products Ltd, Packington, UK. 26

Profile and genetic parameters of dairy cattle lameness NRC 2001. Nutrient requirements of dairy cattle, 7th edition. National Academy Press, Washington, DC, USA. Onyiro OM and Brotherstone S 2008. Genetic analysis of locomotion and associated conformation traits of Holstein-Friesian dairy cows managed in different housing systems. Journal of Dairy Science 91, 322 328. Onyiro OM, Andrews LJ and Brotherstone S 2008. Genetic parameters for digital dermatitis and correlations with locomotion, production, fertility traits and longevity in Holstein Friesian dairy cows. Journal of Dairy Science 91, 4037 4046. Pryce JE, Coffey MP and Brotherstone S 2000. The genetic relationship between calving interval, body condition score and linear type and management traits in registered Holsteins. Journal of Dairy Science 83, 2664 2671. Pryce JE, Nielsen BL, Veerkamp RF and Simm G 1999. Genotype and feeding system effect and interactions for health and fertility traits in dairy cattle. Livestock Production Science 57, 193 201. Rajkondawar PG, Liu M, Dyer RM, Neerchal NK, Tasch U, Lefcourt AM, Erez B and Varner MA 2006. Comparison of models to identify lame cows based on gait and lesion scores and limb movement variables. Journal of Dairy Science 89, 4267 4275. Stevenson JS, Kobayashi Y, Shipka MP and Rauchholz KC 1996. Altering conception of dairy cattle by gonadotropin-releasing hormone preceding luteolysis induced by prostaglandin F2a. Journal of Dairy Science 79, 402 410. Swalve HH, Alkhoder H and Pijl R 2008. Estimates of breeding values for sires based on diagnoses recorded at hoof trimming: Relationships with EBV for conformation traits. Interbull Bulletin 38, 87 90. United States Department of Agriculture 2007. Dairy 2007, Part I: reference of dairy cattle health and management practices in the United States. USDA- APHIS-VS, CEAH, Fort Collins, CO, USA. van der Linde C, de Jong G, Koenen EPC and Eding H 2010. Claw health index for Dutch cattle based on claw trimming and conformation data. Journal of Dairy Science 93, 4883 4891. van der Waaij EH, Holzhauer M, Ellen E, Kamphuis C and de Jong G 2005. Genetic parameters for claw disorders in Dutch dairy cattle and correlations with conformation traits. Journal of Dairy Science 88, 3672 3678. Van Dorp TE, Boettcher P and Schaeffer LR 2004. Genetics of locomotion. Livestock Production Science 90, 247 253. Van Dorp TE, Dekkers JCM, Martin SW and Noordhuizen JPTM 1998. Genetic parameters of health disorders and relationships with 305-day milk yield and conformation traits of registered Holstein cows. Journal of Dairy Science 81, 2264 2270. Whitaker DA, Kelly JM and Smith S 2000. Disposal and disease rates in 340 British dairy herds. The Veterinary Record 146, 363 367. Zink V, Stipkova M and Lassen J 2011. Genetic parameters for female fertility, locomotion, body condition score and linear type traits in Czech Holstein cattle. Journal of Dairy Science 94, 5176 5182. Zwald NR, Weigel KA, Chang YM, Welper RD and Clay JS 2004. Genetic selection for health traits using producer-recorded data. I. Incidence rates, heritability estimates and sire breeding values. Journal of Dairy Science 87, 4287 4294. 27