Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA, USA
Elbow dysplasia (ED) is a skeletal disease that can present as one or several manifestations including fragmented medial coronoid process (FCP) of the ulna, osteochondritis dissecans (OCD) of the humeral condyle, ununited anconeal process (UAP), and incongruity of the elbow joint (INC). It is a complexly inherited disorder, indicating the involvement of multiple genes as well as environmental factors. Environmental variables can include dietary load and degree of activity/mechanical stress.
To understand the genetic background of elbow dysplasia, we have to consider how different genetic factors can contribute to its development. Genetic factors that have been proposed include those controlling endochondral ossification, inherited factors affecting differential growth rates of the three bones of the elbow joint, and liability genes for osteoarthritis. Each individual dog, familial cluster, and breed may have a different mix of genes responsible for the phenotypic expression of ED.
FCP is the most common presentation of ED. Specific incongruities dictate the location of fragmentation or presentation of UAP. These include short radius/long ulna, long radius/short ulna, humeroulnar incongruity, and radioulnar incisure incongruity. Each of these occur due to differences in bone growth rate or growth plate closure. While the OCD lesions seen in ED have been considered and evaluated for primary cartilage disease, many researchers feel that they occur in the specific areas identified due to secondary contact damage, or "kissing lesions".
The different lesions of FCP and UAP vary between breeds, as well as between familial clusters within breeds. Ubbink et al. (1999) found different familial influences in Bernese Mountain Dogs between FCP and INC. However, the specific lesions of FCP were not identified to more accurately represent different causative genotypes. Genomic studies of genes controlling collagen and cartilage formation have not been found to correlate to the development of ED. To date, specific genes affecting differential bone growth or growth plate closure have not been found.
When the frequency of ED is examined between large and small varieties of the same breeds, it is seen with greater frequency in larger varieties than in smaller or miniature varieties. The genetic components of ED may be similarly inherited between these varieties, but without the weight/stress component the pathological lesions may not form. In some breeds, there is an increased frequency of ED in male dogs that appears to be correlated to their increased body size. Some studies show a correlation between the development of ED and hip dysplasia (HD), and some do not. However, it is recognized that both ED and poorer hip conformation can occur more frequently in heavily boned individuals.
Heritability and Penetrance
Heritability (h2) is the phenotypic variation cause by the genotype divided by the total phenotypic variation (the sum of the variation caused by genotype plus the variation caused by environment). It is described as the percentage of variation; e.g., 17% heritable. A completely penetrant monogenetic disorder would be 100% heritable. A higher heritability translates to a greater phenotypic response to selection.
As the computed heritability is directly related to the environmental variation, heritability estimates for genetic disorders will vary between studied populations. A heritability estimate is specific only for that population. The increased the environmental variability that is present, the lower the heritability estimate. Controlling environmental variability produces a higher heritability estimate and perceived inheritance of the disorder.
Published heritability estimates for ED range from 17% to 77%; however, most studies show a heritability of 18% to 21%. This makes elbow dysplasia a moderately heritable disorder, comparable to other complexly inherited traits such as canine hip dysplasia, egg production in poultry, and milk production in cattle. Proper selection against elbow dysplasia should result in a reduction of the frequency of affected dogs. Penetrance is a measure of the percentage of phenotypic expression of a certain genotype. A monogenetic trait that is always expressed is completely penetrant. Most issues of incomplete penetrance have to do with complexly inherited traits where multiple genes must combine to either cross a threshold (qualitative genes) or trigger expression (qualitative genes). Therefore, statements of incomplete penetrance for complexly inherited disease should be more accurately described as the carrying of a genetic load of disease liability genes rather than being genetically affected with incomplete penetrance.
Selection Based on Phenotype
Evaluation of phenotypically normal breeding stock based on a single elbow radiograph under represents the percentage of dog affected with ED. The IEWG recommends multiple views of the elbow for phenotypical diagnosis, and MRI of the elbow is shown to be the most superior imaging modality for ED. The older the age of the dog at imaging, the greater the expression of osteoarthritis secondary to ED.
An important issue with the phenotypical diagnosis of ED is that dogs with Grade I ED have not uniformly been selected against. These dogs rarely show clinical signs of morbidity due to the disorder, but have clearly identifiable radiographic changes. The OFA has shown that having one parent with Grade I ED bred to a normal parent produces twice as much ED (23%) than when two normal parents are used. Whenever a dog with Grade II or III ED is identified, screening of first degree relatives often finds dogs with Grade I ED. Grade I ED is evidence of an accumulation of ED liability genes. Due to this, the BVA/KC has now changed its recommendations to only breed from two parents with radiographically normal elbows.
Using Familial Data for Selection
Complexly inherited disorders show a greater response to selection when it is based on familial data. OFA vertical pedigrees offer a graphical view of depth and breadth of elbow ratings in the pedigree. Familial data can also be computed as estimated breeding values (EBVs), based on the phenotype of the parents, siblings, siblings of parents, offspring, and other relatives. By utilizing phenotypical depth and breadth of pedigree, EBVs utilize information that can more accurately reflect the cumulative genetic influences passed down to the individual dog.
Lewis et al. (2013) computed EBVs for elbow dysplasia on several high registration breeds in the UK Kennel Club registry. They found that the use of EBVs could increase the selection for normal elbows more than 10 fold versus selection based on a single dog's phenotype.
An issue with the accuracy of calculating EBVs involves dogs with missing phenotypes - as most breeds have less than 10% of breeding dogs or their siblings evaluated. In an applied setting, dogs with high EBVs may also become popular sires thus putting pressures on gene pools that can affect genetic diversity.
Genomic breeding values (GBVs) are computed based on the identification of liability genes or quantitative trait loci (QTLs) statistically associated with the disease phenotype. Pfahler and Distl (2012) identified several genetic markers associated with ED in Bernese Mountain Dogs that could be used as a genomic screening panel. However, their work on hip dysplasia has identified genetic liability markers in German Shepherd Dogs and Bernese Mountain Dogs that are different and not shared between the breeds. This shows that genetic marker-based association is specific to the population being studied, and may not reflect genetic liability in a larger or different population.
GBVs and EBVs are continually being improved and enhanced, and will undoubtedly be important in the future control of complexly inherited disorders such as ED. In the meantime, selection based on the best available phenotypical imaging and incorporating familial breadth and depth of pedigree data should improve the elbow status of individual dogs and thus their breeds.
1. Dennis R, Llewellyn A. Change in advice on selecting sires and dams for breeding under the BVA/KC elbow dysplasia scheme. Vet Rec. 2013 Nov 30;173(21):531.
2. Fels L, Marschall Y, Philipp U, Distl O. Multiple loci associated with canine hip dysplasia (CHD) in German shepherd dogs. Mamm Genome. 2014 Jun;25(5–6):262–269.
3. Keller GG, Dziuk E, Bell JS. How the Orthopedic Foundation for Animals (OFA) is tackling inherited disorders in the USA: using hip and elbow dysplasia as examples. Vet J. 2011 Aug;189(2):197–202.
4. Lewis TW, Blott SC, Woolliams JA. Comparative analyses of genetic trends and prospects for selection against hip and elbow dysplasia in 15 UK dog breeds. BMC Genet. 2013 Mar 2;14:16.
5. Michelsen J. Canine elbow dysplasia: aetiopathogenesis and current treatment recommendations. Vet J. 2013 Apr;196(1):12–19.
6. Pfahler S, Distl O. Identification of quantitative trait loci (QTL) for canine hip dysplasia and canine elbow dysplasia in Bernese mountain dogs. PLoS One. 2012;7(11):e49782.
7. Salg KG, Temwitchitr J, Imholz S, Hazewinkel HA, Leegwater PA. Assessment of collagen genes involved in fragmented medial coronoid process development in Labrador Retrievers as determined by affected sibling-pair analysis. Am J Vet Res. 2006;67:1713–1718.
8. Temwichitr J, Leegwater PA, Hazewinkel HA. Fragmented coronoid process in the dog: a heritable disease. Vet J. 2010 Aug;185(2):123–129.
9. Ubbink GJ, Hazewinkel HA, van de Broek J, Rothuizen J. Familial clustering and risk analysis for fragmented coronoid process and elbow joint incongruity in Bernese Mountain Dogs in The Netherlands. Am J Vet Res. 1999 Sep;60(9):1082–1087.