Scientific Basis for More Views and More Care for Overinterpretation
World Small Animal Veterinary Association World Congress Proceedings, 2014
Anu Lappalainen, DVM, PhD
Department of Equine and Small Animal Medicine, University of Helsinki, Helsinki, Finland

Elbow dysplasia (ED) refers to four hereditary developmental disorders of the elbow joint: the medial coronoid process disease (MCPD), also known as FCP (fragmented coronoid process), osteochondrosis (OC) of the medial part of the humeral condyle, ununited anconeal process, and incongruity of the elbow joint. Medial compartment disease (MCD) refers to all pathologies (MCPD, OC and the contact or "kissing" lesion caused by friction of the diseased MCP) of the medial side of the elbow joint. Each type can present alone or in any combination. Incongruity is currently proposed to be a common aetiology for the other three forms of ED.

Some elbow conditions are not considered to be included in ED (e.g., calcified bodies seen near or at the medial epicondyle of the humerus). Calcified bodies have been referred to as ununited medial epicondyle, medial epicondylar spur, or calcified body in the joint capsule. In recent literature, these conditions are grouped together, as they are all new bone formation of tendons, either at the insertion or further away of the joint,1 and the term flexor enthesopathy has been suggested. In Labrador retrievers, a hereditary background has been suggested.2

Medial coronoid process disease can be a diagnostic challenge radiographically since the radiological findings are often sparse or non-existent.3 Therefore, an approach using secondary osteophytes as an indicator of ED was advocated in screening protocols.4 However, abnormal contour and indistinctive cranial border of the MCP and subtrochlear sclerosis are radiological signs frequently seen in mediolateral projections (ML) in MCPD. Subtrochlear sclerosis, manifesting as increased radiopacity at the base of the coronoid process visible in ML radiographs, has proved to be a reliable indicator of MCPD.

Osteochondrosis is radiographically best diagnosed from craniocaudal (CrCd) oblique radiographs as a radiolucent defect at the articular surface of the medial humeral condyle. A differential diagnosis for OC on radiographs is an abrasive contact ("kissing") lesion caused by MCPD.

Computed tomography (CT) is considered an accurate method for imaging the canine elbow joint, although some lesions might not be visible. CT can in most cases clearly show MCPD.

Comparison of Radiographic and CT Findings of Belgian Shepherd Dogs and Labrador Retrievers with Grade ED1 Elbows5,6

Belgian shepherd dogs and Labrador retrievers have fairly similar screening statistics according to the Finnish Kennel Club (FKC)'s database (12% and 17%, respectively), but Belgian shepherd dogs seldom have clinical elbow disease compared to Labrador retrievers. In Finland, ED1 refers to mild OA with osteophyte formation of < 2 mm detected usually on the dorsal surface of the anconeal process (Table 1).

Table 1. Elbow dysplasia grades and their definitions used in the Finnish screening protocol

Grade 0 (free)

No signs of OA.

Grade 1 (mild)

Mild OA with osteophyte formation of < 2 mm detected usually on the dorsal surface of the anconeal process.

Grade 2 (moderate)

Osteophytes on the dorsal surface of the anconeal process 2–5 mm high, changes in the MCP or the joint is mildly deformed.

Grade 3 (severe)

Marked degenerative changes visible, or osteophyte formation on the anconeal process is over 5 mm high, UAP.

OA = osteoarthritis; MCP = medial coronoid process; UAP = ununited anconeal process

In the studied population, according to the CT, MCPD was evident in 14% of the joints in 19% of the Belgian shepherd dogs (Figure 1). In contrast, MCD was found in 54% of the joints in 77% of the Labrador retrievers (Figure 2).

Figure 1. Radiographic grading of the 36 joints with and without medial coronoid process disease based on computed tomography in 18 Belgian shepherd dogs
Figure 1. Radiographic grading of the 36 joints with and without medial coronoid process disease based on computed tomography in 18 Belgian shepherd dogs

MCPD = medial coronoid process disease, 0b = borderline dysplasia.

Figure 2. Radiographic grading of the 26 joints with and without medial compartment disease based on computed tomography in 13 Labrador retrievers
Figure 2. Radiographic grading of the 26 joints with and without medial compartment disease based on computed tomography in 13 Labrador retrievers

MCD = medial compartment disease.

The accuracy of the FKC grading of mild ED differed clearly between the breeds (Table 2). In Belgian shepherd dogs, grading based on the FKC protocol (new bone formation on the dorsal border of the anconeal process) was unreliable, yielding high percentages of both false-positives and false- negatives. However, in Labrador retrievers the protocol proved to be accurate in grading dogs as dysplastic or non-dysplastic (ED0 versus ED1). In Belgian shepherd dogs, blurring of the cranial edge of the MCP and subtrochlear sclerosis were reliable signs of MCPD. In Labrador retrievers, the accuracy of radiographic signs indicative of MCPD was good and equivalent to the accuracy of grading. In Finland, evaluation of screening radiographs has recently changed to some extent, with grading focusing more on radiographic signs indicative of MCPD.

Table 2. Sensitivity and specificity (%) for radiographic findings and grading of the 36 elbow joints in 18 Belgian shepherd dogs and of the 26 joints in 13 Labrador retrievers


Belgian shepherd dogs

Labrador retrievers










Osteophytes CRCD view





Bony opacity on AP





Subtrochlear sclerosis





Blurring of MCP





MCP = medial coronoid process; AP = anconeal process; CrCd = craniocaudal projection; N/A = not available
* p ≤ 0.05

In the CrCd oblique view, a radiolucent subchondral bone defect on the medial humeral condyle was observed in 19% and flexor enthesopathy in 4% of the joints in Labrador retrievers. In this breed, the CrCd oblique view would have assisted in correct diagnosis of MCD in 12% of the cases. Sensitivity and specificity of osteophytes seen on the CrCd oblique radiograph were 93% and 92%, respectively (Table 2). A CrCd view alone would have given a definite diagnosis of MCD in 19% of the joints, as a subchondral bone defect in the medial humeral condyle was observed giving increased confidence in grading the joint as dysplastic with ED grade 3 instead of ED grade 1. Additionally, the CrCd oblique view increased reliability of assessing OA.

Radiographic screening for ED from one ML flexed radiograph is common in many countries. Screening in all Nordic countries has mostly been based on secondary lesions, the main reason for this being difficulty in detecting primary lesions from radiographs, which are often of suboptimal technical quality. The use of secondary lesions in a screening protocol where the minimum age is commonly 12 months is based on the assumption that osteophytes develop in juvenile animals. However, often only minimal or no degenerative changes are visible on radiographs, as was the case also in our study. On the other hand, the secondary changes are visible on the radiographs of many breeds, like Labrador retrievers, and these changes are a good indicator of MCD in these breeds.

Osteophytes arise usually first on the proximal surface of the anconeal process. In our study, it was impossible with certainty to differentiate osteophytes from normal anatomic variation of the anconeal process in Belgian shepherd dogs using CT. However, differentiating between osteophytes and normal anatomy of the anconeal process might have less weight in evaluation of the screening radiographs in Belgian shepherd dogs if the radiological findings of MCPD weigh more in grading.

Flexor enthesopathy is often not visible in ML radiographs, but the bone fragment is clearly visible in the CrCd view. This condition is not considered a part of ED, but since it can be a cause of lameness1 and is thought to be hereditary in Labrador retrievers,2 collecting data as part of the screening protocol would be beneficial. Without a CrCd view, a substantial proportion of these lesions could be missed. On the other hand, CrCd projection was not helpful in screening for ED in Belgian shepherd dogs, again emphasizing differences between breeds.


1.  de Bakker E, Gielen I, Saunders JH, Polis I, Vermeire S, Peremans K, Dewulf J, van Bree H, van Ryssen B. Primary and concomitant flexor enthesopathy of the canine elbow. Veterinary and Comparative Orthopaedics and Traumatology. 2013;Epub Jun 26.

2.  Paster ER, Biery DN, Lawler DF, Evans RH, Kealy, RD, Gregor TP, McKelvie PJ, Smith GK. Un-united medial epicondyle of the humerus: radiographic prevalence and association with elbow osteoarthritis in a cohort of labrador retrievers. Veterinary Surgery. 2009;38:169–172.

3.  Goldhammer MA, Smith SH, Fitzpatrick N, Clements DN. A comparison of radiographic, arthroscopic and histological measures of articular pathology in the canine elbow joint. Veterinary Journal. 2010;186:96–103.

4.  Grøndalen J. Arthrosis with special reference to the elbow joint of young rapidly growing dogs. II. Occurrence, clinical and radiographical findings. Nordisk Veterinärmedicin. 1979;31:69–75.

5.  Lappalainen AK, Mölsä S, Liman A, Laitinen-Vapaavuori O, Snellman M. Radiographic and computed tomography findings in Belgian shepherd dogs with mild elbow dysplasia. Veterinary Radiology & Ultrasound. 2009;50:364–369.

6.  Lappalainen AK, Mölsä S, Liman A, Snellman M, Laitinen-Vapaavuori O. Evaluation of accuracy of the Finnish elbow dysplasia screening protocol in Labrador retrievers. Journal of Small Animal Practice. 2013;54:195–200.


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

Anu Lappalainen, DVM, PhD
Department of Equine and Small Animal Medicine
University of Helsinki
Helsinki, Finland

MAIN : IEWG – Elbow Dysplasia in Dogs : More Views & Overinterpretation
Powered By VIN