REQUEST PERSONAL ACCOUNT
You can get access to the Knowledge Base by requesting your personal account.
REQUEST 30 DAY FREE TRIAL ACCOUNT
In order to try out and check the benefits of Vets-net.com you can get a free trial account.
STUDENTS CREATE ACCOUNT HERE
Students can request their account by selecting their associated college.

News

04-12-2008 - Prevalence of hip dysplasia, elbow dysplasia and humeral head osteochondrosis in dog breeds in Belgi

Prevalence of hip dysplasia, elbow dysplasia and humeral head osteochondrosis in dog breeds in Belgium

F. Coopman, DVM, PhD1, G. Verhoeven, DVM1, J. Saunders, DVM, PhD1, L. Duchateau, IM, PhD2 and H. van Bree, DVM, PhD1

1 Department of Medical Imaging, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium
2 Department of Physiology and Biometrics, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium

The official screening results of the Belgian National Committee for Inherited Skeletal Disorders, an affiliate of the Belgian Kennel Club, have been used to estimate the prevalence of hip dysplasia, elbow dysplasia and humeral head osteochondrosis in the dog breeds in Belgium, and these have been compared with reported prevalence data from other countries. In some breeds, the prevalence of hip and elbow dysplasia is very high, both in Belgium and in other countries. Comparisons of the prevalence of hip dysplasia are not always feasible because different systems are used to evaluate the quality of the hips and because there is no strict consensus on what should be considered a diseased hip joint.

HIP dysplasia was first reported in dogs by Schnelle (1935). It is now known that it is a quantitatively inherited trait (Leppänen and Saloniemi 1999), with estimates of its heritability ranging from 0·17 to 0·6 (Fisher 1979, Lingaas and Klemetsdaal 1990, Tomlinson and McLaughlin 1996, Swenson and others 1997a). Dysplastic dogs have a higher risk of producing affected offspring (Hedhammar and others 1979, Keller 2006).

Several methods are used to screen dogs for the condition. The Pennsylvania Hip Improvement Program (PennHIP) system tests the hip capsule for passive laxity (PennHIP 2007). The methods of the Fédération Cynologique Internationale (FCI), the Orthopedic Foundation for Animals (OFA) and the British Veterinary Association/Kennel Club (BVA/KC) evaluate osseous conformation and evidence of osteoarthritis, but also signs of subluxation. Other methods that test functional subluxation are the dorsolateral subluxation projection and the method of Flückiger (Todhunter and others 2003b).

In the FCI system, hips that are scored C, D or E are considered to be dysplastic (Brass 1993). In the OFA system, dysplastic hips are described as mild, moderate or severe (Keller 2006), and all other cases are considered to be unaffected. This indicates that in the FCI and OFA systems a cut-off value is described clearly, but not followed consistently (Verhoeven and others 2007). A score of 11 or more given by the BVA/KC evaluation panel indicates that the hip is either grossly unstable or has clearly established secondary changes (Gibbs 1997), but it is not stated whether this score demarcates between hips that are affected and unaffected. Wood and others (2000, 2002) showed that, in this system, the scores for the two hips of a dog are very similar, so that if a dog has one hip with a score of more than 10, its other hip is likely to have a score of more than 10. This indicates that dogs with a total score of more than 20 have either grossly unstable hips, or hips with clearly established secondary changes. In the PennHIP system (Smith 2000), dogs with a distraction index (DI) of 0·28 (German shepherd), 0·32 (golden retriever), 0·38 (labrador retriever) or 0·35 (rottweiller) or less are considered to be free of the condition (Smith and others 2001). Dogs with a DI greater than these thresholds have a higher risk of developing degenerative joint disease in the hip joint, that is, confirmed hip dysplasia. A DI above which all dogs would develop degenerative joint disease has not been described, except for German shepherd dogs in which a DI above 0·9 is predictive for the development of the disease (Smith and others 2001). The dorsolateral subluxation (DLS) projection, described by Farese and others (1998), scores hips in terms of the percentage of the femoral head covered by the acetabulum, and combines hip joint laxity and hip conformation. Dogs with DLS scores of more than 50 per cent are considered to have a low probability of developing degenerative joint disease, a score between 45 and 55 per cent is correlated with a medium probability, and a score of less than 45 per cent is correlated with a high probability (Todhunter and others 2003a). Flückiger and others (1999) consider that the use of the subluxation index (SI) gives identical predictions to the DI. Hips with a SI of 0·3 or less have a low probability of being classified as dysplastic, hips with a SI between 0·3 and 0·5 have a greater probability, and 95 per cent of hips with a SI greater than 0·5 are classified as dysplastic (Flückiger and others 1999).

The term elbow dysplasia was introduced by Corley and Carlson (1965). It is a combination of four different disorders (Samoy and others 2006), which, according to the International Elbow Working Group (IEWG) (Flückiger 2000) are primary defects. Osteoarthritis and osteophytes are described as secondary lesions. The disease is known to be partly inherited in many dog breeds (h2 = 0·25 to 0·28), (Beuing and others 2000), and a breeding programme to control it has been successful in rottweilers and Bernese mountain dogs (Swenson and others 1997b, Audell 2000). In the Scandinavian countries and in the USA, dogs are diagnosed with the disease if they are positive for the secondary lesions, whereas in Germany, the Netherlands, Switzerland, Belgium and France, dogs with only a primary defect may also be diagnosed with the disease.

Humeral head osteochondrosis in dogs was described by Schnelle (1954), Brass (1956), Vaughan (1968) and Smith and Stowater (1975); the labrador retriever (Morgan and others 1999) and 24 other pure breeds, and mixed breeds (LaFond and others 2002) have been reported to be susceptible. Morgan and others (1999) described the disease as a heritable joint disorder, and LaFond and others (2002) concluded that a breeding programme aimed at controlling the disease in the breeds at risk might be worthwhile. There is no evidence that different methods are used to evaluate humeral head osteochondrosis.

The objectives of this study were, first to report the prevalence of hip dysplasia, elbow dysplasia and humeral head osteochondrosis in the dog breeds in Belgium that were officially screened between January 1, 2002 and June 15, 2006, and secondly to compare these data with prevalence data from other countries.

MATERIALS AND METHODS

Official evaluation procedure
Between January 1, 2002 and June 15, 2006, radiographs of the hips, elbows and shoulders of potential breeding dogs were sent to the Belgian National Committee for Inherited Skeletal Disorders (NCISD) for official evaluation. Hips were evaluated according to the criteria of the FCI, as described by Brass (1993). Elbows were evaluated according the IEWG criteria, as described by Flückiger (2000). Shoulder joints that had signs of osteochondrosis all received grade 1. A sign of humeral head osteochondrosis is when the humeral head is not round. After extensive discussions between the panellists, it was decided that a dog with obvious signs of osteoarthritis, but no signs of osteochondrosis, should not be considered as a positive case, because other causes of osteoarthritis, such as trauma or medial glenohumeral ligament problems, could not be ruled out. The dog's owners were informed that there were signs of osteoarthritis, but the dog was considered negative for humeral head osteochondrosis.

All the evaluations were by three observers (board-certified by the European College of Veterinary Diagnostic Imaging and observers with more than 10 years' experience). If no consensus between these observers was found, a board-certified radiologist with 30 years' experience was asked to make the final decision.

Data collection and processing
The evaluations, together with their age, sex and breed were collected from all the dogs in the NCISD database. The scores for hip and elbow dysplasia were transformed as follows. For hips, grades A and B were considered to be unaffected, grade C mildly affected, and grades D and E moderately to severely affected. For elbows, grade 0 was considered to be unaffected, grade 1 mildly affected and grades 2 and 3 moderately to severely affected.

The overall prevalence of the conditions in all the dogs, regardless of their breed, and specific prevalences for breeds having 20 or more evaluations were estimated on the basis of these data.

To compare the Belgian prevalence data with prevalence data from other countries, the literature was searched for reports on the overall and breed-specific prevalences of the three conditions in other countries. In the case of the BVA/KC and PennHIP data, the prevalences were estimated by assuming that the scores for the disorders were normally distributed. A dog is considered to be radiographically affected if the hip is scored as C, D or E (FCI), if the hip scores 21 or more (BVA/KC), or if it has a DI or 0·7 or more (PennHIP). The OFA reports simply in terms of whether the hip is affected or unaffected.

RESULTS

Summary statistics
Radiographs of dogs of 174 different breeds were sent to the NCSID for official screening. For hip dysplasia, 5883 dogs were given a score, for elbow dysplasia 1356 dogs were given a score, and for humeral head osteochondrosis, 267 dogs were given a score. The median and mean ages of the dogs were 560 days and 698 days, respectively, and 60 per cent of them were female. The youngest dog was one year of age and the oldest nine years. Only 3 per cent were more than five years old, and 71·4 per cent were less than two years old when the radiographs were taken.

Prevalence of the three conditions in dog breeds in Belgium
The overall prevalence, and the prevalence in specific breeds, of hip dysplasia, elbow dysplasia and humeral head osteochondrosis in Belgium are listed in Table 1, 2 and 3, respectively. Overall, one in five dogs had hip dysplasia. Four popular dog breeds in Belgium, the German shepherd dog, golden and labrador retriever and the Bernese mountain dog had almost one in four affected individuals. None of the 40 breeds in Belgium mentioned in Table 1 could be considered free of hip dysplasia. Elbow dysplasia seemed to be a greater problem in rottweilers, Newfoundlands and Chinese shar peis (Table 2). Nevertheless, the Bernese mountain dog and golden retriever, both among the breeds with a high prevalence of hip dysplasia, also had a high prevalence of elbow dysplasia. Humeral head osteochondrosis affected fewer dogs (Table 3).


View this table:
[in this window]
[in a new window]

TABLE 1 : Overall prevalence and breed-specific prevalence of hip dysplasia in dogs in Belgium (January 1, 2002 to June 15, 2006)



View this table:
[in this window]
[in a new window]

TABLE 2 : Overall prevalence and breed-specific prevalence of elbow dysplasia in dogs in Belgium (January 1, 2002 to June 15, 2006)



View this table:
[in this window]
[in a new window]

TABLE 3 : Overall prevalence and breed-specific prevalence of humeral head osteochondrosis in dogs in Belgium (January 1, 2002 to June 15, 2006)


Prevalence of the three conditions in breeding dogs in other countries
Flückiger and others (1995) recorded a prevalence of 42 per cent for hip dysplasia in all breeds between 1970 and 1994. The OFA (Corley 2000) estimated that its prevalence in all breeds was 17 per cent in 1980 and 12 per cent in 1994. Rettenmaier and others (2002) found that 18·2 per cent of dogs had signs of hip dysplasia irrespective of whether they had a pedigree.

Tables 4 and 5 list the breed-specific prevalences of hip dysplasia in dogs in different countries. The prevalences in Table 4 are based on data from this study, and studies by Swenson and others (1997a), Leppänen and Saloniemi (1999), Morgan and others (1999), Corley (2000), Lawson (2000), Mäki and others (2000, 2001), Wood and others (2000, 2002), Hamann and others (2003), Smith (2005) and the OFA (2006b). The prevalences in Table 5 are based on data from this study, and studies by Morgan and others (1999), Corley (2000), Lawson (2000), Smith (2005) and the OFA (2006b). The prevalence of hip dysplasia in the four most popular breeds in Belgium seems to be comparable with the prevalence in the USA (OFA and PennHIP) and Sweden. In contrast, the prevalence of hip dysplasia (mild, moderate and severe combined) in Finland is up to 10 per cent higher. Leppänen and Saloniemi (1999) found the prevalence of severe hip dysplasia to be 16 to 18 per cent in the boxer, 4 to 7 per cent in the flat-coated retriever, 18 to 21 per cent in the golden retriever, 14 to 19 per cent in the Irish setter, 17 to 22 per cent in the labrador retriever and 20 to 31 per cent in the rottweiler. This is higher than the prevalence of moderate to severe hip dysplasia in the same breeds in Belgium (Table 1).


View this table:
[in this window]
[in a new window]

TABLE 4 : Percentage breed-specific prevalence of hip dysplasia in dogs in different countries in specific time periods



View this table:
[in this window]
[in a new window]

TABLE 5 : Percentage breed-specific prevalence of hip dysplasia in dogs in different countries in specific time periods


Table 6 lists the breed-specific prevalences of elbow dysplasia in dogs in different countries. The prevalences are based on data from this study, and studies by Morgan and others (1999), Audell (2000), Beuing and others (2000), Mäki and others (2000, 2001), Remy and others (2004) and the OFA (2006a). The prevalence of elbow dysplasia found in rottweilers in Belgium is the lowest of the countries compared. In Germany and Sweden, prevalences of up to 60 to 65 per cent have been recorded. In the USA, the Newfoundland, Chinese shar pei and Bernese mountain dog have high prevalences of elbow dysplasia similar to the high prevalence found in Belgium.


View this table:
[in this window]
[in a new window]

TABLE 6 : Percentage prevalence of elbow dysplasia in dogs in different countries in specific time periods


Data on the prevalence of humeral head osteochondrosis was available only for a population of labrador retrievers examined at the School of Veterinary Medicine, University of California, Davis, USA in which the prevalence in the late 1990s was estimated as 3·7 per cent (Morgan and others 1999).

DISCUSSION

The data summarised in Tables 1, 2, 3 achieve the first objective of this study. The comparison of the Belgian prevalence data with the prevalence data from other countries is not straightforward because of the different screening systems used. Furthermore, in some systems, there is no clear cut-off value for deciding whether a dog is affected. Prevalence data on elbow dysplasia will differ because in some countries primary defects are not evaluated, whereas in others they are considered to be positive cases. Differences in the age when the dogs are screened, and the sex ratio in different countries also have to be considered when prevalence data are compared.

Only dogs with hips with a score of A in the FCI system, excellent in the OFA system, 0 in the BVA/KC system, or a DI less than 0·28 can strictly be considered radiographically free of hip dysplasia. In comparison, a hip joint can reasonably be considered dysplastic when its score is more than 20 in the BVA/KC system or C, D or E in the FCI system, when it is at least mildly affected in the OFA system, or when its DI is 0·7 or more in the PennHIP system. Owing to this comparability, the cut-off values applied in this study can be used to make an equal radiographic distinction between affected and unaffected hips, although the distinction has not been validated. As a result, the Belgian data can be compared with data from other countries that use different systems.

The dogs that are scored in the USA by the OFA system are at least two years of age (Keller 2006), older than most of the dogs scored in this study, in other countries using the FCI system (Swenson and others 1997a, b, Beuing and others 2000) and in the study by Morgan and others (1999). Dogs scored by the PennHIP system are much younger. Owing to these differences in the age at which the dogs were screened, it is not possible to compare the prevalence of hip dysplasia at a standardised age. This should not pose too serious a problem because in this study only cases with gross instability and secondary changes have been considered, rather than cases with only secondary changes. This implies that although in older dogs more cases of degenerative joint disease will be found, the total numbers of dogs with hip dysplasia will not differ consistently, because the cases with gross instability are also considered. On the other hand, the systems that use the hip-extended method (FCI and OFA), or animals that are too young (FCI), may record some false negative results because degenerative joint disease is not yet apparent and passive laxity is masked (Smith 2000, Keller 2006). As a result, the Belgian data may be biased towards an underestimate in comparison with some other countries.

As in most other studies, there were more female than male dogs (Morgan and others 1999, Beuing and others 2000, Wood and others 2000, 2002), and as a result the estimated prevalence may be biased. However, when comparing the prevalence with that in other countries any bias will be small because the preponderance of females is general.

Another problem in comparing the different prevalences is the period over which they are being compared. It can be expected that the prevalence will be lower if it is being determined on the basis of more recent data, because of the selection towards sound hips and elbows, although Willis (1997) reported that certain breeds have not shown a decrease in prevalence. Differences in the number of animals considered may also cause problems.

Despite these restrictions, the results indicate that the prevalence of the three conditions in dogs in Belgium is, to some extent, comparable with their prevalence in other countries. When comparing the data in this study with those reported by Leppänen and Saloniemi (1999), Mäki and others (2000) and Paster and others (2005), the prevalence in this study seems to be underestimated, probably because many radiographs are not officially recorded (Tomlinson and McLaughlin 1996, Smith 1998, Rettenmaier and others 2002). Mäki and others (2000) stated explicitly that all the radiographs of dogs that are screened in Finland are sent for official evaluation, but this is not the case in Belgium. The prevalences of mild, moderate and severe hip dysplasia observed in Finland may therefore give a more accurate estimate than the Belgian data. In particular, the fact that the prevalence of severe hip dysplasia is much higher in Finland than in Belgium suggests that, in Belgium, severe cases may not be sent for evaluation, because they are very clear and therefore often not sent for official evaluation by practitioners.

Whatever the differences, and whatever the reasons for them may be, the prevalence of hip dysplasia and elbow dysplasia is high if zero tolerance for such genetic disorders is considered for dogs in Belgium and other countries. The conditions remain a problem in Belgium, other European countries and the USA, and stringent measures will be needed to reduce the prevalence of these diseases to acceptable levels.


< Back to news