The Organization of an ED-Screening Program and the Use of the Certificate
World Small Animal Veterinary Association World Congress Proceedings, 2014
H.A.W. Hazewinkel, DVM, PhD, DECVS, DECVCN
Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands


Elbow dysplasias (ED) occur frequently in 3–5-month-old dogs of medium and large body size, during the period of high growth velocity. It is known that each form of ED will lead to osteoarthrosis (OA) with possibly severe consequences for the well-being of the animal and its owner. First, definitions will be given of the entities of ED followed by screening-organizing aspects.


Primary Lesions (1–4)

In the screening programme according to the International Elbow Working Group (IEWG), these lesions are graded as absent (ED grade 0), suspected-present (ED grade II), or present (ED grade III). We will distinguish the following primary lesions:

1.  UAP: Separation in the cartilaginous bridge between the secondary ossification centre of the anconeal process and the olecranon, which can cause a partially or completely detached anconeal process, referred to as ununited anconeal process (UAP).

2.  FCP or MCPD (= medial coronoid process disease): Fissuring of the medial coronoid process of the ulna with partial to complete separation (fragmentation) of the medial coronoid process from the ulna; primary subchondral bone lesions with secondary articular cartilage changes,1 although also chondromalacia at the medial coronoid process is considered part of this entity.

3.  OC or OCD: Local thickening of growing epiphyseal cartilage with delayed endochondral ossification, which may develop into OCD with a single or fragmented detached cartilage flap.
"Kissing lesion"= An abrasion of the articular cartilage, sometimes extending into the subchondral bone (radiologically often slightly more lateral than the OC lesion), and here caused by a fragmented coronoid process.1 This finding is graded as a "OCD-like lesion."

4.  Elbow incongruity (EI, INC): The subchondral bone of the trochlear notch of the ulna and of the radial head are not parallel to the opposing humeral subchondral bone. There are different forms of EI:

a.  The radius longer than the ulna with a narrowing of the joint space between the tip of the anconeal process and the humeral condyle, a distally gradual widening of the joint space between the ulnar semilunar notch and the humeral condyle and the radial head proximal of the coronoid process of the ulna.

b.  The longer ulna with a wider joint space between the proximal radius and the humeral condyle and the step between the more proximally located distal edge of the ulnar trochlear notch (i.e., the lateral coronoid process) and the radial head (and displacement of the distal humerus cranially).2 This can also be considered as an underdeveloped or too small trochlear notch.

c.  The alignment between the subchondral bone of the trochlear notch and the radial head is more elliptical than the circular contour of the humeral condyles described by Wind.3

d.  Developmental elbow luxation with lateral displacement of the (often hypoplastic) radial head with a comparative overgrowth of the radius (as seen in case of chondrodysplasia in non-chondrodystrophic breeds).

Osteoarthrosis (OA) is radiologically characterized by new bone formation at the edges of the joint. In addition, enthesophytes (i.e., new bone formation at the sites of attachments of tendons, ligaments, and joint capsule, resulting from abnormal tension placed on the soft-tissue attachments near the joint margins) can be formed. Regardless of the primary cause, the pattern of OA is similar. The different locations where osteophytes and enthesophytes are visible in case of OA are given in Figure 1.

Figure 1. Locations for grading of elbow OA according to IEWG
Figure 1. Locations for grading of elbow OA according to IEWG

a. The proximal surface of the anconeal process; b. The cranial aspect of the radial head; c. The cranial edge of the medial coronoid process; d. The caudal surface of the lateral condylar ridge; e. Sclerosis of the ulnar notch, in the medulla; f. On the medial surface of the medial epicondyle; g. At the medial edge of the medial coronoid process; h. Indentation of the subchondral bone: OCD (-like) lesion; i. Spur formation (enthesophyte, no OA).

Sclerosis. An alteration in normal bone architecture (i.e., a decrease in normal bone porosity) is depicted on a ML view of the elbow joint as an increase of bony opacity with loss of trabecular markings (a white area), just caudal to the lateral coronoid process in the ulna. Sclerosis is considered as one of the first signs of ED in young dogs, especially when the primary cause cannot be identified, as in some cases of FCP. This area can be compared with a control radiograph of the non-affected elbow in case of unilateral ED. However, since FCP often occurs bilaterally, the use of the opposite elbow joint will not be of help. In a survey with 17 Labrador retrievers (6–16 months of age) with FCP and 17 without FCP as diagnosed by arthroscopy, radiographic density was objectivated and expressed as pixels: an extremely significant correlation between pixel intensity of the projection of the lateral coronoid process revealed in dogs with FCP.4 However, in a study in Labradors of two age groups (first lameness <12 months of age or >12 months of age) the sclerosis was remarkably absent in the dogs with first lameness at an older age [Theyse: Different Presentations of Medial Coronoid Disease at Different Ages: A Clinical, Radiological, CT, and Arthroscopic Study].5 Microscopically, this intramedullary area is characterised by reduced intertrabeculae spaces either due to mechanical overloading or influence of MMPs, enzymes which play a role in osteoarthrosis. Ulnar osteotomy may resolve the sclerosis indicating overloading as important cause.

Additional findings include periarticular mineralisation (mineralisation/avulsion of flexor tendons at the medial epicondyle), OA of unknown origin, or any other abnormality noted should preferably be reported as well. These abnormalities might have various aetiologies and variable relevance (and can be registered for future research purposes).

Grading of Elbow Osteoarthrosis (OA) according to IEWG

 Grade 0 OA: no signs of osteophytosis or osteosclerosis

 Grade I OA: When at any of the locations listed a–i osteophytes are present of <2 mm, or presence of osteosclerosis

 Grade II OA: When at any of the locations listed a–i osteophytes are present of 2–5 mm

 Grade III OA: When at any of the locations listed a–i osteophytes are present of ≥5 mm

Borderline OA can be defined as increased radiographic density (sclerosis) in the ulna caudal to the trochlear notch. In addition, minimal changes at the dorsal border of the anconeal process which is considered as a normal edge and grouped under borderline. This can be scored separately or as Grade 1.

In several countries the presence of a primary lesion such as UAP, FCP, OCD, or INC of >2 mm, automatically results in an ED score 3; the suspicion of primary lesions results in an ED score 2.

Imaging Techniques

Radiographs play a major role in the diagnosis of ED, both in a clinical setting, as well as in screening the population. "More views will give more insight" counts also true in case of radiological investigation, especially in case the primary lesion is of importance to know. According to a large study in 447 Bernese Mountain Dogs by Lang et al.,6 12% of the dogs had a primary ED without OA yet. Therefore, screening for ED in Bernese Mountain dogs should include at least two perpendicular views. This seems especially true in breeds where OCD is anticipated to be the primary cause of ED (Lappalainen: Scientific Basis for More Views and More Care for Overinterpretation). In case the secondary signs only are of importance, a limited number of views can be sufficient. In addition to the ML and AP views, other views have been developed including ML view with 15-degree supination (exorotation) of the antebrachium or distomedial-proximolateral oblique view.7

Increase in quality (correct positioning, exposure, and contrast) can compensate for the quantity of radiographs; in a practical setting not all radiographs which are made available can be included in the screening procedure. More views increase the costs, and depending on the national legislation, more positions may complicate and expand the radiological session. More frequently, computed tomography (CT) is used instead or in addition of radiography. Dogs have to be anesthetized and positioned in dorsal recumbency on the scanner table with the forelimbs extended and the antebrachia parallel to each other during scanning (Figure 2). Transverse views are made perpendicular to the antebrachia in 1-mm thick slices with 120 kV, 120 mA, and 1-sec scanning time. Series are reviewed in transverse slices, as well as in sagittal and dorsal reconstructions. The evaluation criteria included the following signs: periarticular osteophytosis; abnormal contour and structure of the medial coronoid process; cyst-like lesion at the radial incisure of the ulna and humeral trochlea; irregularity of the radial incisure of the ulna; and evidence of other primary lesions, such as fragmentation of medial coronoid process, radioulnar incongruity, and OCD-like lesions.5,8

Figure 2. Labrador positioned in dorsal recumbency on the computed tomographic scanning table
Figure 2. Labrador positioned in dorsal recumbency on the computed tomographic scanning table

See Reference 5.


The breeders' club has to decide which animals must be screened: at least the dogs used for breeding should be screened before allowance to breed is obtained. Only dogs free from ED should be used, so ED grade 1–3 must be excluded. Including several dogs of each litter in the screening program allows for calculating the estimated breeding value of the members of the litters and thus supports the breeding program strongly.9

The dogs should have a mature skeleton since growth plates may cause artefacts. The dog should have had the chance of developing signs of OA. Therefore the dog must have a minimal age of 12 months (IEWG) or 24 months (OFA). For practical reasons, often the minimal age (as set by FCI) for HD screening will be adopted for the ED screening. Although for HD screening sedation is prescribed by FCI, sedation or anesthesia will not influence the ED scoring. Sedation will however facilitate positioning of the front legs with the aid of sand bags, in case the authorities do not allow manual positioning.9

The radiographs should be sent to the screening authority, often the breeders' club or the national kennel club, where the administration (including identification of the dog) related to the screening is performed. A panel of screeners, experienced and independent of the breeders' club and owners, should screen the (blinded) radiographs according to a strict protocol and report this to the administration who will provide the owner with a certificate.

Example of ED certificate
Example of ED certificate

From page 27.

An appeal procedure should be possible both for owners/breeders and for practitioners when radiographs are rejected because they did not fulfil the criteria as set by the kennel club, or when there is doubt about the degree of OA/ED as declared in the certificate. The appeal panel should be independent from the first panel; it should be stated beforehand if there is or is not always the need for a new set of radiographs, or there is always the need for another technique (e.g., a larger set of radiographs, or CT scanning). And it should be stated beforehand that the last judging counts. Especially in case of ED grade I, there may be the wish to reevaluate the dog: for example, reevaluation is allowed on a new set of radiographs not within 12 months after the first screening took place.9

More important than a well-organized screening procedure, is the conclusions the breeders will draw from the fact that the breed is affected with a hereditary disease, the incidence of which can only gradually decrease when there is full cooperation of the breeders, even the incidental breeders (i.e., correct offspring registration), no prescreening but central registration of all affected animals (operated or not), and exclusion of animals at risk.


1.  Morgan JP, Wind A, Davidson AP, eds. Hereditary Bone and Joint Diseases in the Dog: Osteochondroses, Hip Dysplasia, Elbow Dysplasia. Hannover, Germany: Schlütersche GmbH & Co. KG, Verlag und Druckerei; 2000.

2.  Samoy Y, Van Ryssen B, Gielen I, Walschot N, van Bree H. Review of the literature: elbow incongruity in the dog. Vet Comp Orthop Traumatol. 2006;19:1–8.

3.  Wind A. Elbow incongruity and developmental elbow diseases in the dog. J Am Vet Med Assoc. 1986;22:711–724.

4.  Burton NJ, Comerford EJ, Bailey M, Pead MJ, Owen MR. Digital analysis of ulnar trochlear notch sclerosis in Labrador retrievers. J Small Anim Pract. 2007;48:220–224.

5.  Lau Seng Fong. Development of medial coronoid disease in Labrador retrievers: diagnostic and pathogenic studies. Thesis Utrecht University, October 2013, ISBN 978-90-393-5962-5964.

6.  Lang J, Busato A, Baumgartner D, Flückiger M, Weber UT. Comparison of two classification protocols in the evaluation of elbow dysplasia in the dog. J Small Anim Pract. 1998;39:169–174.

7.  Haudiquet PR, Marcellin-Little DJ, Stebbins ME. Use of the distomedial-proximolateral oblique radiographic view of the elbow joint for examination of the medial coronoid process in dogs. Am J Vet Res. 2002;63:1000–1005.

8.  Reichle JK, Park RD, Bahr AM. Computed tomographic findings of dogs with cubital joint lameness. Veterinary Radiology & Ultrasound. 2000;41:125–130.

9.  IEWG 2012 proceedings (


Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

H.A.W. Hazewinkel, DVM, PhD, DECVS, DECVCN
Faculty of Veterinary Medicine, Utrecht University
Utrecht, The Netherlands


B. Tellhelm
Department of Small Animal Clinics, Section of Surgery
University of Giessen