Breeding Against Hip and Elbow Dysplasia in Dogs

The objective of this study was to develop breeding strategies to reduce the incidence of canine hip and elbow dysplasia. For this purpose, investigations were conducted on the mode of inheritance as well as genetic parameters, genetic trends and selection differentials for the traits. Population structure and inbreeding were also examined, and finally, various selection schemes were compared by a simulation study to find out potential genetic responses and possibilities for a comprehensive breeding program. Screening information for hip and elbow dysplasia, collected within the health program of the Finnish Kennel Club, was used. A total of 42 421 records on hip dysplasia and 13 543 on elbow dysplasia were available on seven breeds, the German Shepherd, the Golden Retriever, the Labrador Retriever, the Rottweiler, the Bernese Mountain Dog, the Finnish Hound and the Rough Collie. Based on the results of this thesis, both hip and elbow dysplasia seemed to be quantitative traits. Neither of the traits fitted to a mitochondrial or sex-linked inheritance. According to segregation analysis, a major gene could be segregating in hip dysplasia in four studied breeds. Estimates of heritabilities in the studied breeds indicated that there is sufficient genetic variation for successful selection of breeding dogs according to their dysplasia status. The positive, although mainly low, estimates of genetic correlations between hip and elbow dysplasia allow for simultaneous selection for both traits. In spite of this, the genetic level of hip dysplasia had improved over the study period only in the Rottweiler, while the genetic levels were getting worse in the other three studied breeds. For elbow dysplasia, a slight improvement was noticed in all studied breeds. However, the phenotypic selection differentials for both hip and elbow dysplasia were quite small in all studied breeds. Despite of the weak selection practised against hip and elbow dysplasia, the populations studied in this thesis were observed to be otherwise under stringent selection, since only a small proportion of annually born animals were used later for breeding. In addition to the small proportion of breeding animals, a few sires had been excessively favoured in breeding. Besides passive inbreeding due to the resulting small effective population sizes, inbreeding had also been used actively in the studied breeds. The inbreeding coefficient of a dog had a significant detrimental effect for hip dysplasia in the German Shepherd and the Labrador Retriever. No inbreeding depression for elbow dysplasia was observed. However, the amount of pedigree information used in this thesis was not necessarily sufficient to properly study inbreeding depression. Potential genetic responses for hip and elbow dysplasia were studied by simulation. When simultaneously selecting for hip and elbow dysplasia, as well as behaviour and appearance, a very small relative index weight for appearance was enough to maintain the trait without negative genetic gain. The only way to reach the breeding goal set for hip and elbow dysplasia was to clearly favour the traits in the selection. Behaviour traits should receive also much more emphasis than has been given to them so far. The most effective selection against hip and elbow dysplasia would be based on BLUP breeding values. However, the accuracy of the breeding values could be improved if a larger proportion of the dogs were screened for hip and elbow dysplasia. Further, if major genes affecting hip and elbow dysplasia exist, considerable genetic progress could be achieved by selection for them, since the possibly existing unfavourable alleles were found to have relatively high frequencies. Systematic breeding programs, including preservation of genetic variation, should be developed and followed to maintain or improve the desired behaviour and working traits, as well as the health of the dogs in the current populations.