Breeding against osteochondrosis : phenotypic and genetic analyses in horses and pigs

Osteochondrosis (OC) is a disturbance in the process of endochondral ossification during skeletal growth. The objectives of this thesis were to assess the prevalence and genetic parameters of OC, and to facilitate optimization of breeding against OC in horses and pigs. In the Dutch warm blood horse, 70% of individuals showed joint abnormalities in at least one of the 28 locations examined. Joint abnormalities were divided into flattened bone contours and bone fragments. Genetic parameters were estimated for flattened bone contours, fragments and overall OC. Heritability was 0.08 for flattened bone contours, 0.22 for fragments, and 0.23 for overall OC. In fattening pigs, the prevalence of OC was 41%, and 12% of the individuals had severe OC. The prevalence was highest for individuals kept on a concrete, partially slatted floor with ad libitum feeding (58%), and lowest for individuals kept on a deep litter floor with restricted feeding (34%). These results demonstrate that the prevalence of OC can be reduced by applying deep litter floors and restricted feeding. The differences in growth patterns between pigs with and without OC were investigated also. After 28 days of age, piglets with severe OC at slaughter started to grow faster, and were significantly heavier after 70 days of age than pigs without OC at slaughter. These results suggest that OC might be related to high growth rates during a specific time period. To evaluate prospects for breeding against OC, selection responses were compared between breeding schemes using phenotypic selection based on own performance or progeny testing and genomic selection. The results show that, when genomic information has an equal accuracy as phenotypic information before selection, it will have a lower equilibrium response to selection because it has a larger Bulmer effect. Especially for low heritable traits, genomic selection was of additional value over traditional selection. The general discussion argues that OC has both economical and practical relevance for the current horse industry. Simulations showed potential for genomic selection compared to the current breeding program in the Dutch warmblood horse (KWPN), even with small reference populations. In pigs, OC is a more serious problem in sows than in fattening pigs, because sows live longer. A decrease in OC prevalence will, therefore, yield larger economic and welfare benefits then apparent from results on fattening pigs. Also for pigs, simulations indicated possibilities for genomic selection. The final section of the general discussion proposes directions for future research.