The association between hoof lesions in early lactation and culling risk in Jersey cows

Introduction 
Lameness is an important indicator of dairy cattle welfare and has a large economic impact due to its impact on milk production and culling. Our specific objectives were to evaluate the association between the presence of early lactation hoof lesions and culling and to determine the association between hoof lesions that develop between 20-120 DIM and culling. 

Materials and Methods 
Cows from this study were selected from the transition facility of a 9000 cow commercial Jersey farm as part of another study. Sampling strategy for the original study consisted of a convenience sample of cows coming of the rotary parlor when the fresh groups were milked. Enrolment criteria selected no concurrent health problems that were 20 ± 3 DIM (d20) postpartum. At enrolment cows were sorted, sampled for blood parameters and given a body condition score (BCS) and all 4 feet were examined for hoof lesions in a hydraulic hoof trimming chute. Lesions were classified as sole hemorrhage (SH), sole ulcer (SU), white line disease (WLD), digital dermatitis (DD), footrot, toe, and non-hoof lesions. No horn was removed from the sole to detect lesions. Cows with hoof lesions other than SH were treated as per farm protocol. At 120 ± 3 DIM (d120), cows were reevaluated for HL and BCS. Survival analysis using Cox proportional hazards regression was used to determine the association with hoof lesions and culling. Cows were censored at the time of culling or when they reached 300 DIM in their enrolment lactation. Two models were constructed 1) all lesions grouped into 1 category, 2) lesion grouped into hoof horn (SU, WLD, toe), SH, infectious (DD, footrot) and non-hoof.

Results 
A total of 1545 cows were included in this study. The majority, 907 (58.7%) of enrolled cows were in their 1^st lactation. At d20, 414 (26.8%) cows had foot lesions with SH (19%) and SU (4.7%) being the most frequent foot lesions. Prevalence of DD and WLD was 1.8% and 1.6% respectively. A total of 267 (17.3%) cows were culled. Median DIM at culling was 149. In the Cox proportional hazards model with all lesions grouped together the hazard of culling was 1.14 (95% CI 0.88-1.49) for cows with lesions compared to cows without lesions. From a model that categorized lesions into categories, the hazard ratios were 1.33 (95% CI 0.93-1.91) for hoof horn, 0.91 (95% 0.67-1.24) for SH, 1.5 (95%CI 0.86-2.60) for infectious lesions and 0.92 (95%CI 0.29- 2.93) for non-hoof lesions. 

Conclusions 
The prevalence of SH in our study was higher while the prevalence of SU, DD and WLD was much lower than expected. The majority of our estimates for hazard ratio indicate a wide range of plausible associations and include both a decreased hazard and an increased hazard. For example cows with hoof horn lesions were observed to have an 33% higher rate of being culled. However, this estimate ranges from cows with hoof horn lesions having a 7% lower culling rate up to a 91% higher culling rate. There are a variety of reasons for our wide range of plausible associations including lower than expected lesion occurrence, treatment of lesion at the time of enrolment and a relatively short follow up period post lesion identification. Our data suggests that the prevention of cows with hoof horn lesions could reduce culling hazard. In addition, high occurrence of SH in early lactation warrants further long term investigation.