Methods

Results

This study included a total of 17,861 radiographic records from 13 dog breeds, including the American Akita, Alaskan Malamute, Old German Shepherd Dog (Altdeutscher Schäferhund), American Staffordshire Terrier, Australian Shepherd Dog, Belgian Shepherd Dog (including Malinois, Groenendael, Laekenois, and Tervuren), White Swiss Shepherd Dog, Bernese Mountain Dog, Cane Corso, Czechoslovakian Wolfdog, Rhodesian Ridgeback, Rottweiler, and Dogue de Bordeaux. The overall prevalence of ED in our study population was 11.4%, ranging among breeds from 1.1% in the Czechoslovakian Wolfdog to 32.2% in the Dogue de Bordeaux. The breeds with the highest prevalence of ED were the Dogue de Bordeaux (32.2%), Rottweiler (21.3%), Bernese Mountain Dog (19.7%), and Cane Corso (15.7%). The ED prevalence for all 13 evaluated breeds is shown (Table 2). In our study population, mild disease (ED1) was most commonly diagnosed in all but 1 breed, the Dogue de Bordeaux, for which severe disease (ED3) was most prevalent. The proportion of dogs screened in this study relative to the number of births was comparatively low, varying from 0.2% for the American Staffordshire Terrier breed to 7.8% for the Dogue de Bordeaux breed. Data were not available for the Old German Shepherd Dog as there is no club for this breed.

Five breeds had more than 200 cases of ED. Interestingly, a primary lesion was identified in 661 dogs among these 5 breeds, which represented a prevalence of 37.4% in the total dysplastic population. The percentage of dogs with ED that had a radiographically visible primary lesion varied across different breeds: 11.9% for the Rottweiler, 25.1% for the Australian Shepherd Dog, and 25.9% for the White Swiss Shepherd Dog. In the Dogue de Bordeaux and Bernese Mountain Dog breeds, the prevalence of primary lesions in cases of ED was much higher (68.8% and 72.7%, respectively) (Table 3).

We then assessed the prevalence of ED in male and female dogs, with and without ED, in the total study population and in the 5 most commonly affected breeds (Table 4). In the total study population, ED prevalence was statistically higher in males, compared with females (17.5% vs 10.5%, P < .01). Similarly, in the Australian Shepherd, White Swiss Shepherd, Rottweiler, and Dogue de Bordeaux breeds, male dogs showed a higher prevalence of ED, compared with female dogs. This difference was statistically significant for the Australian Shepherd Dog (P = .043), Rottweiler (P < .01), and Dogue de Bordeaux breeds (P < .01). In the Bernese Mountain Dogs, the prevalence of ED was similar in males and females (P = .882).

During the study period, the overall ED prevalence decreased as the number of dogs screened increased (Figure 1). During the period 2002 to 2010, a significant decrease in ED prevalence was calculated in the overall number of evaluated dogs with a negative correlation coefficient of 0.7 (P = .043). Over the period 2011 to 2022, no correlation was found.

View Full Size

Evolution of the prevalence of elbow dysplasia (ED) and the number of screened dogs between 2002 and 2022.

Citation: American Journal of Veterinary Research 85, 6; 10.2460/ajvr.23.12.0290

• Download Figure
• Download figure as PowerPoint slide

Discussion

In this study, we report a descriptive retrospective analysis of radiographic ED screening data in dogs in France over a 20-year time period (2002 to 2022). We found an overall prevalence of 11.4% in the total study population, with the lowest prevalence in the Czechoslovakian Wolfdog (1.1%) and the highest prevalence in the Dogue de Bordeaux (32.2%). A primary lesion was identified in 37.4% of dogs with ED. Elbow dysplasia prevalence was statistically higher in male dogs (17.5%), compared with female dogs (10.5%). Moreover, we showed that joint incongruity and FCP were the most common primary lesions in dogs with ED and that most dogs had mild (ED1) disease. A significant decrease in ED prevalence was reported in the overall number of evaluated dogs over the study period.

Our data are consistent with previous reports showing that ED prevalence ranges from 0% to 55% depending on the breed and screening program.^1,13–16 Several factors may influence data from official screening programs. First, many countries have mandatory ED screening, which provides more reliable data due to the higher number of screened dogs relative to the total number of dogs in each breed. Furthermore, breeding methods may influence data, since a more rigorous selection of breeding animals is expected to reduce the prevalence of dysplasia in that breed. Finally, various screening methods are used in different countries. In the US, Finland, Sweden, and Italy, only a single ML radiographic view with a flexed elbow (the minimum required by the IEWG) is required for ED screening, which may lead to underdiagnosis of primary lesions.^2,8,10,17,18

In this study, the Dogue de Bordeaux, Rottweiler, Bernese Mountain Dog, and Cane Corso breeds were most commonly affected by ED, with prevalences of 32.2%, 21.3%, 19.7%, and 15.7%, respectively. These results reflect the recognized predisposition of large, rapidly growing dogs to develop ED.^1,16,18 In the Rottweiler and Bernese Mountain Dog breeds, our prevalences (21.3% and 19.7%) were similar to those of the official mandatory ED surveillance program in Sweden, which found prevalences of 26.2% and 17.4%, respectively, in those 2 breeds (official data between 2001 and 2021 published by the Svenska Kennelklubben).^19,20 In the Dogue de Bordeaux breed, the previously reported prevalences ranged from 25.5% (US) to 47% (Finland) and were consistent with our results (32.2%).^21,22 These data highlight the high prevalence of ED in this breed and the importance of selective breeding. Moreover, the Dogue de Bordeaux was the only breed in which severe disease (ED3) was diagnosed in the majority of ED cases included in this study. The analysis of primary lesions must be correlated with the age of the animals during screening, as the radiographic lesions of a dysplastic dog tend to be more severe with age.^18,23–25The majority of dogs screened in this study were between 12 and 13 months old for 12 breeds, except for the Dogue de Bordeaux, which had an age of around 15 months. Screening was performed at the discretion of the breeder or owner and was conducted in young animals, generally in their second year of life.

In the Cane Corso breed, our prevalence (15.7%) was lower than those reported in the literature.^{15,16,20–22} However, official screening is currently optional for this breed in France, whereas it is mandatory for registering a Cane Corso litter in Sweden. Moreover, there were only 383 radiographs of this breed in our study, compared with 1,350 in Sweden and 2,180 in the US over a similar time period.^20,22 Our prevalences for the American Staffordshire Terrier, Rhodesian Ridgeback, and Belgian Shepherd Dog were also lower than published data.^20,21 As these breeds are subject to official screening in Sweden (all 3 breeds) and Finland (Rhodesian Ridgeback), the larger number of screened dogs may have enabled a more accurate evaluation of prevalence in those countries.^20,21 Indeed, mandatory screenings allow for obtaining a true prevalence of ED in the breed, avoiding any possible preselection during nonmandatory screening, based on clinical signs or radiographic images of the dog.^11,13 For the White Swiss Shepherd Dog breed, the ED prevalence in our study was lower than that observed in other countries. In most of these countries, the number of screened dogs was lower than in our study (3,144 readings). However, over our study time period (2002 to 2022), only 7.5% of White Swiss Shepherd Dogs were screened for ED in France; therefore, our results may underestimate the real prevalence of ED. Four other breeds in our study (American Akita, Alaskan Malamute, Australian Shepherd Dog, and Czechoslovakian Wolfdog) showed a low prevalence of ED (< 4%), which is consistent with previous reports.^20–22

In accordance with previous findings, we found that the ED prevalence was higher in male dogs than in female dogs in the total study population and in Australian Shepherds, Rottweilers, and Dogues de Bordeaux.^6,24,26 A sex-related predisposition to ED has been well established in these 3 breeds.^16,27 Hormone levels, rapid growth, and weight gain may play a role in this predisposition and some risk alleles may be on the X chromosome.^17,23,25 In contrast, we found no difference in ED prevalence between male and female White Swiss Shepherd dogs, which may be due to the relatively small number of dogs screened for this breed. The ED prevalence was also similar in male and female Bernese Mountain Dogs.

In our study, primary ED lesions were not commonly identified in the total study population. The rate of identification of primary lesions was lower than a previous study, which reported that only 4% of the total population and 1.8% of Bernese Mountain Dogs diagnosed with dysplasia had osteoarthritis without primary lesions.²⁵ However, that study included suspected primary lesions, whereas our results were based on identified primary lesions. We also found that the rate of identification of primary lesions varied according to previous studies, due to the difference in pathogenesis of ED between breeds, resulting in different primary lesions and presumably depending on the size of the bone and age of the dog.²⁵

Our data showed that FCP and joint incongruity were the 2 most common primary lesions in the 5 breeds with the most cases of ED. In the Bernese Mountain Dogs, more than 50% of dogs with ED had joint incongruity and almost half of Dogues de Bordeaux (44.4%) with ED had an FCP. FCP is the most frequent primary lesion in dysplastic dogs while previous reports^{1,5,7,9,25,28} showed a radiographic sensitivity ranging from 10% to 62% based on osteophytosis or the visibility of fragments. Differences between screening methods and the number of radiographs required may affect the identification of FCP and other primary lesions.^2,12 Several studies^1,14 have suggested a causal relation between joint incongruity and the development of other lesions, notably OCD and FCP. Moreover, the predisposition of Bernese Mountain Dogs and Rottweilers for FCP, with or without joint incongruity, has been well established. Osteochondritis dissecans and UAP were infrequently identified in this study, similar to previous reports. This may be because obvious OCD and UAP lesions can be easily identified on radiographs, and therefore, an unknown proportion of cases may have been identified by veterinarians and thus not submitted to the screening program.

Previous studies²⁹ reported that phenotypic selection over time reduced ED prevalence in Rottweilers and Bernese Mountain Dogs. In Germany, the proportion of unaffected Rottweilers rose from 34.8% to 53.8% over 4 years (1996 to 1999).²⁹ In Sweden, the ED prevalence fell from 52% in 1985 to 24% in 1997 in Bernese Mountain Dogs and from 57% in 1984 to 37% in 1997 in Rottweilers.^26,27 In our study, we found that ED prevalence significantly decreased during the period 2002 to 2010 in the overall number of dogs evaluated suggesting a positive impact of radiographic screening on reducing ED prevalence. Care should be taken when selecting on the basis of physical characteristics alone, as phenotypic signs of dysplasia may not be present. The cause of ED is not completely understood, but the disease is considered to be multifactorial. Environmental factors, such as diet and the type and amount of exercise, contribute to disease development in genetically predisposed individuals.^{1,3,17–19,23,29,30} ED is a polygenic disease with incomplete penetrance. As a result, not all genetically predisposed individuals exhibit dysplasia.

Complementary diagnostic methods can improve the effectiveness of phenotypic selection to reduce ED.^15,18,23 Various factors, including age at screening and the use of sedation during screening, can affect the results of radiographic screening.^10,18,24 Therefore, the radiographic screening result for an individual dog may not be sufficiently accurate for selection purposes. While genetic testing is the technique of choice to identify breed-related disorders, it is not currently available for ED, a polygenic disease for which all causative genetic mutations are not known. Complementary information can be provided using the best linear unbiased prediction (BLUP) method, which combines all available phenotypic information from relatives in a pedigree while simultaneously adjusting for environmental effects.^18,19,23 By using BLUP, the genetic gain per generation improves more than by simple phenotypic selection, but BLUP tends to increase the inbreeding rate per generation.^18,19,23

The main limitation of our study is the relatively small number of individuals tested for ED, compared with the number of registered births for each breed over the study period. Analysis of the ED prevalence over time is subject to a number of biases, including variations in the number of readings per year and the motivations prompting for screening as it is not mandatory. Furthermore, this study did not include ED screening of other predisposed breeds, such as German Shepherds or Labrador Retrievers as the 2 examiners (JPG and DF) were not official readers for these breeds. Consequently, our results cannot be extrapolated to the entire population of dogs in France. Moreover, the screened population may be skewed toward breeding dogs, as breeders are generally more aware of screening for hereditary diseases than owners. Therefore, our data should be interpreted with caution, and further studies are needed to evaluate the effectiveness of screening programs in France.

Disclosures

The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.

Funding

The research was funded by the authors’ departments.