Oertzenweg 19 b
+49 30 838 62356
Elbow dysplasia (ED) is the most common cause of forelimb lameness in large dog breeds. It
results in non-responsive, painful osteoarthritis, even though it has been treated surgically,
with NSAIDs and/or by physiotherapy from the very first. This is why since the 70ies there
have been many approaches to develop an elbow replacement, as it was successfully done
for the hip replacement. To date this has not worked out.
Objective of the study was to assess the morphometrics of the elbow of ED predisposed
large dog breeds by using computed tomography (CT). This data should be compared for the
development of a new total elbow replacement.
First, the procedure was validated on macerated bones and defined as gold standard with
regard to recumbencies and measurements. Then this gold standard was used for stored
CT-data of clinical patients, that were gained due to diagnostics of forelimb lameness without
following a standardized protocol.
Macerated bones: 16 elbows of 8 dogs of large (mixed-) breeds were used for validation. To
define the gold standard of this study the elbow joint was released from soft tissue,
macerated and dried. Then humerus and ulna were sawed at 7 user-defined distances and
measured at the cutting edges by a calliper. Possible errors in measurement because of
maceration and drying (p = 0,07) were evaluated as non-significant. The intra-observer
failure was documented as ± 0,2mm by repeated measurements.
Previously the same 7 distances were measured CT-based on the elbow joint still being in
the cadaver of those dogs which were positioned with elbows in 90° flexion and any
recumbency of the forelimb. The macerated bones were measured isolated in the CT as well.
Results: The measurements of all three methods were compared to the gold standard and
coincide with those by a tolerated error of ± 0,4 mm. Therewith every single CT-based
measurement is precisely enough for determining morphometrics of bones. Hence, all stored
clinical data of patient’s joints can be used for morphometrics, which means there is a lot
more data volume to use.
Patients: 44 elbows of dogs of ED-predisposed large breeds of stored clinical CT-data of the
small animal clinic of the FU Berlin were used to determine the morphometrics of the elbow.
Dogs where excluded because of any invasive joint trauma, suspected tumors, especially
osteosarcoma, or generalized skeletal disease. The elbows of Labrador Retriever suffering
from ED showed mild differences to those without ED (e. g. the humeral condyle was 12 %
longer), which would not have an influence on the dimensions of an elbow replacement.
Results: The elbow shows bilateral symmetry to the axis of the body (Student T-Test on
variation Art. Cubiti dexter et sinister between p = 0,08 and p = 1). It is becoming bigger by the factor 1,75: Labrador Retriever < Golden Retriever < Rottweiler < Berner Sennenhund <
Deutscher Schäferhund < Bordeauxdogge. There are no changes in the anatomoical shape.
Only the humeral condyle becomes 10 % more compact, when the joint becomes bigger.
The elbow joint is mainly isometrical. Only the humeral condyle differs from that, because of
its conical shape with a circular base, that is medial 3 mm larger than lateral.
Conclusion: With the results of the morphometrics and measured parameters it is possible,
to define a medium, minimum and maximum joint for any dog of ED-predisposed large
breeds and to give recommendations for a new 2-component total elbow replacement
consisting of a humeral and a radioulnar part.
The humeral part is mounted into the medullary canal by one stem, the radioulnar part by two
- one for the radius and one for the ulna. Near to the articulating surfaces the medullary canal
is wider than distal, this is why the stem should be conical as well. The stem of the humeral
portion should be elliptical near to the humeral condyle minimum 13,8 mm x 2,7 mm and
maximum 28,2 mm x 11,2 mm and distal minimum 5,3 mm x 7,2 mm and maximum
14,9 mm x 22,4 mm. The radial stem should be a conical cylinder with an expanse of
minimum 3,1 mm x 5,5 mm and maximum 11,6 mm x 17,4 mm (proximal), which becomes
35 % smaller at the distal end. The ulnar stem is a cylinder with mean expanse of
8,2 mm (± 2,3 mm) x 6,6 mm (± 2,4 mm).
The humeral head portion replaces the humeral condyle. Due to its morphometrics it should
be conical (s. above) with a minimum length of 33,8 mm and a maximum length of 50,7 mm.
The humeral condyle is symmetric to its minimum radius, which divides the condyle into half.
This is the point were to insert the stem to the head portion.
The radioulnar head portion should be fixed concerning the angle between Inc. trochlearis
and the axis of the medullary canal with 153° (± 5,4°). The radioulnar artiuclating surface is
the product of width by length of the Fovea capitis radii, minimum 1,25 cm2 to 3,21 cm2 and
the length between Proc. coronoideus lat. to med. 16,7 mm to 23,1 mm and the depth of the
Inc. trochlearis 3,8 mm to 7,8 mm.