The Veterinary Journal 186 (2010) 96–103

Contents lists available at ScienceDirect

The Veterinary Journal

journal homepage: www.elsevier.com/locate/tvjl

A comparison of radiographic, arthroscopic and histological measures

of articular pathology in the canine elbow joint
Marc A. Goldhammer a,*, Sionagh H. Smith b, Noel Fitzpatrick c, Dylan N. Clements d
a
Tierärztliche Klinik Dr. Schneider-Haiss, Karl-Heinrich Käferlestr. 2, 71460 Ludwigsburg, Germany
b
Royal (Dick) School of Veterinary Studies, Division of Veterinary Clinical Sciences, The University of Edinburgh, Veterinary Pathology Unit, Easter Bush Veterinary Centre,
Roslin EH25 9RG, UK
c
Fitzpatrick Referrals, Halfway Lane, Eashing, Godalming, Surrey GU7 2QQ, UK
d
Royal (Dick) School of Veterinary Studies, Division of Veterinary Clinical Sciences, The University of Edinburgh, Hospital for Small Animals, Easter Bush Veterinary Centre,
Roslin EH25 9RG, UK

a r t i c l e i n f o a b s t r a c t

Article history: Validation of radiographic and arthroscopic scoring of joint pathology requires their comparison with his-
Accepted 28 July 2009 tological measures of disease from the same joint. Fragmentation of the medial coronoid process (FMCP)
is a naturally occurring disease of the canine elbow joint that results in osteoarthritis, and the objectives
of this study were to compare the severity of histopathological changes in the medial coronoid process
Keywords: (MCP) and medial articular synovial membrane with gross radiographic scoring of elbow joint osteophy-
Canine tosis and the arthroscopic assessment of the MCP articular cartilage surface.
Fragmentation of the medial coronoid
Radiographic scoring of osteophytosis and the arthroscopic scoring of visual cartilage pathology of the
process
Histopathology
MCP correlated moderately well with the histopathological evaluation of cartilage damage on the MCP
Synovium and synovial inflammation in the medial part of the joint, but not with bone pathology in the MCP.
Cartilage islands Marked cartilage pathology on the MCP was identified in joints with either no radiographic evidence
of osteophytosis or with mild cartilage damage that was evident arthroscopically.
Ó 2009 Elsevier Ltd. All rights reserved.

Introduction MCP disease (Carpenter et al., 1993). Osteoarthritis is characterised

Fragmentation of the medial coronoid process of the ulna
(FMCP) is a common developmental disease of the canine elbow
joint most frequently observed in young medium to large breed
dogs (Mason et al., 1980; Meyer-Lindenberg et al., 2002; Moores
et al., 2008). The aetiology of FMCP is undetermined at present
(Danielson et al., 2006; Haudiquet and Rochereau, 2007). Several
factors have been implicated in the past, such as radioulnar joint
incongruity (Wind, 1986; Wind and Packard, 1986; Samoy et al.,
2006; Gemmill and Clements, 2007), trauma (Guthrie et al.,
1992; Haudiquet and Rochereau, 2007), genetics (Ekman and Carl-
son, 1998), growth rate (Ekman and Carlson, 1998), nutrition,
ischaemia (Ekman and Carlson, 1998) and osteochondrosis (Grond-
alen and Grondalen, 1981; Wolschrijn et al., 2005). Direct identifi-
cation of MCP disease, as well as the assessment of the articular
surface, can be achieved by arthroscopic (van Ryssen et al., 1993;
Bardet, 1997; van Ryssen and van Bree, 1997; van Bree and van
Ryssen, 1998) or open arthrotomy evaluation of the elbow joint.
Osteophytosis is the radiographic hallmark of osteoarthritis
(OA), and is often the only radiographic finding associated with
* Corresponding author. Tel.: +49 7141 86888.
E-mail address: goldhammermarc@hotmail.com (M.A. Goldhammer).
by destruction of articular cartilage, but the involvement of other
tissues in the pathogenesis of the disease process such as bone
(Rogers et al., 2004), synovium (Dingle, 1981), fat (Ushiyama
et al., 2003) and ligaments (Hill et al., 2005) is recognised. The
radiographic changes which characterise OA are delayed relative
to histological (Gilbertson, 1975) and gene expression changes
(Stoker et al., 2006) in experimental canine OA. The histological
changes reported with MCP disease in dogs include fatigue micro-
damage of the bone, increased bone porosity, a reduction in osteo-
cyte density, and thickening of articular cartilage (Danielson et al.,
2006). Changes in gene expression have been identified in cartilage
and bone of the diseased MCP which correlate with the radio-
graphic grade of OA (Clements et al., 2009). However, the utility
of clinical measures as an indicator for disease severity at a micro-
scopic level has not been evaluated in naturally occurring canine
joint disease.
In this study we evaluated the clinical measures of osteoarthri-
tis (radiographic and arthroscopic changes) in elbows affected by
MCP disease and related them to the histological changes observed
on the surgically removed MCP. All findings were compared to a
control group of dogs without evidence of elbow osteoarthritis or
MCP disease. We hypothesised that arthroscopic changes of the
articular cartilage surface of the MCP would be closely related to

1090-0233/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tvjl.2009.07.025
 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103
the histological changes observed in cartilage of the MCP, but not
to the bone of the MCP or the synovium of the medial aspect of
the joint. We further hypothesised that these changes would not
correlate with the severity of the radiographic changes identified
on each elbow joint.
Material and methods
Samples
The MCP was evaluated by arthroscopy using a 2.7 mm 30° oblique arthroscope
(Hopkins) in 53 elbows of 34 dogs [Labrador (22), cross breed (3), German Shepherd
(2), Golden Retriever (2), Rottweiler (2), Mastiff (1), Newfoundland (1), Hungarian
Viszla (1)] with lameness and elbow discomfort attributable to FMCP. Fragmenta-
tion of the medial coronoid process was diagnosed by arthroscopic and arthrotomy
evaluation, as previously described (Danielson et al., 2006). The mean age of the af-
fected dogs was 11 months (±4.5 months, range 5–23 months) with a mean body-
weight of 30.7 kg (±12.1 kg, range 7.2–70.0 kg).
Orthogonal radiographs were obtained from each elbow. The visual changes
identified in the articular cartilage were scored at the time of surgery (Modified
Outerbridge Score; Schulz, 2003) by a single experienced observer (NF). Subtotal
coronoid ostectomy was performed by removing the MCP at its base perpendicular
to its long axis by using an oscillating saw (Danielson et al., 2006). Synovial mem-
brane was taken from the medial aspect of the joint at the arthrotomy incision in
cases (n = 40) where resection of the redundant joint capsule was required for accu-
rate closure of the medial arthrotomy.
Control samples (MCP and synovium) were collected from 10 dogs that were
euthanased for reasons other than fore limb lameness. Five Beagles, two cross-
breeds, two Labradors and one German Shepherd dog were evaluated. None showed
clinical or radiographic evidence of elbow disease. In each case no evidence of el-
bow disease was evident upon visual inspection of the joint. The mean age of the
control dogs was 17 months (±19.5 months, range 9 months–6 years) and the mean
weight of the control dogs was 22.7 kg (±7.8 kg, range 13.2–40.1 kg). Tissue samples
were washed in sterile Hartmann’s solution (Isolec, Ivex Pharmaceuticals) and
immediately stored in 10% neutral buffered formalin until processing.
The MCP samples were decalcified for up to 48 h in Decalcifier I (Surgipath). All
samples were then paraffin wax embedded and sectioned at 5 lm thickness. Sec-
tions of the MCP were cut perpendicular to the weight bearing surface of the coro-
noid process, through the apex of the process to the midpoint of the ostectomy
surface. If loose fragmentation was present, the MCP fragment was evaluated,
and sectioned to the midpoint of its base. Two serial sections were mounted onto
glass slides and stained with haematoxylin and eosin (HE) and Safranin-O (SO).
All synovial membrane samples were stained with HE only.
Radiographic scoring
The orthogonal views of the affected elbows were evaluated for the presence of
osteophytosis and were scored accordingly by a single experienced observer (NF)
using the International Elbow Working Group Method (IEWG) (Grade 0, no osteo-
phytes; Grade 1, osteophytes <2 mm; Grade 2, osteophytes 2–5 mm; Grade 3,
osteophytes >5 mm) (Lang et al., 1998).
Arthroscopic scoring
The visual appearance of the articular cartilage across the surface of the MCP
(but not including the fragmentation line) identified during arthroscopy was scored
using the modified Outerbridge cartilage scoring system (Schulz, 2003): Grade I,
signs of chondromalacia, namely, softening and swelling; Grade II, partial thickness
fibrillation and fissuring of the cartilage; Grade III, full-thickness cartilage fissuring;
Grade IV, full-thickness cartilage erosion with exposure of the subchondral bone.
Histopathological evaluation
The sectioned coronoid processes were evaluated microscopically (Olympus BX
41), and digital images (Olympus, CAMEDIA) were obtained concurrently for further
analyses.
Images of each sample were digitally captured at 200 magnification and ana-
lysed using ImageProPlus software (Media Cybernetics). Cartilage thickness was
measured at three different sites perpendicular to the weight bearing surface and
a mean of the three measurements was calculated (performed by MG). In Outer-
bridge Grade IV cases, where full-thickness cartilage erosion had been identified,
the three measurements were taken adjacent to the denuded bone.
Each section was evaluated by using the Mankin histological and histochemical
scoring system (Mankin et al., 1971; Bobinac et al., 2003) more recently referred to
as the Histologic Histochemical Grading System (HHGS) (Ostergaard et al., 1997,
1999). This enabled semi-quantitative evaluation of structural changes in all layers
97
of the uncalcified cartilage, the tidemark integrity and the level of SO stain uptake.
Reduced SO uptake generally equates to a reduction in proteoglycan content in the
cartilage matrix (Jubb and Eggert, 1981).
The sections were evaluated at different magnifications for the individual parts
of the scoring procedure (performed by MG and SS). The Mankin cartilage score was
calculated for all of the samples. According to the Mankin grading system, cartilage
degeneration was categorised into three stages: Stage I, Mankin score 0–6 points –
indicates mild degenerative changes; Stage II, Mankin score 7–9 points – indicates
moderate degenerative changes; Stage III, Mankin score 10–14 points – indicates
severe degenerative changes (Mankin et al., 1971).
The synovial samples were assessed histologically using a scoring system de-
scribed by (Krenn et al., 2006). Sections were assessed microscopically for thicken-
ing (i.e. hyperplasia) of the synovial lining cell layer, the density of resident cells
and for the presence of infiltrating inflammatory cells (Table 1). A score of 0–1 point
was considered normal, while a score of 2–4 points was classified as low-grade
synovitis and 5–9 points as high-grade synovitis (Krenn et al., 2006).
Three digital images at different sites of the section were obtained at 400
magnification. The images were imported into ImageProPlus software, and an area
of interest (AOI,) was defined by choosing an area immediately subjacent to the
tidemark boundary between articular cartilage and subchondral bone without over-
lap between the three areas. The number of osteocyte nuclei was counted in the AOI
and the osteocyte density (number of osteocytes per 80,000 lm2) was calculated.
Using ImageProPlus software, the AOI was measured and its bone surface area
calculated automatically by outlining the bony trabeculae with a digital marker,
which determined the bone area in lm2. Three measurements were taken and
the mean was calculated and expressed in per cent compared to the whole area
(80,000 lm2) on which the measurements were based.
During the scoring process, additional microscopic abnormalities not encom-
passed by the semi-quantitative scoring systems or by the image analysis were re-
corded qualitatively for the synovium, articular cartilage and subchondral bone.
Statistical analyses
The age and weight of the dogs in the diseased and the control group were sta-
tistically compared using Student’s t test. Statistical analyses to compare the IEWG
elbow dysplasia score (Lang et al., 1998), the Outerbridge score (Schulz, 2003) with
the mean articular cartilage thickness, mean osteocyte density, mean bone area, the
Mankin cartilage pathology score, and the synovial scores were performed using
one-way analysis of variance with a post hoc Tukey’s pair-wise comparison. Differ-
ences were considered significant at P < 0.05. Correlation between the severity of
the IEWG elbow dysplasia score, or the Outerbridge gross pathologic findings of
the diseased samples when compared to the mean articular cartilage thickness,
mean osteocyte density, mean bone area, Mankin cartilage pathology score, and
the synovial scores was performed by calculating the Pearson correlation co-
efficient.
Results
When compared to the diseased population by t test, the control
dogs were significantly lighter (P = 0.04) but not significantly older
(P = 0.14).
Radiographic evaluation of the elbows
The evaluation of the radiographs revealed a distribution of the
IEWG grades in the evaluated elbows of: Grade 0 (n = 15), Grade 1
(n = 11), Grade 2 (n = 14), Grade 3 (n = 13).
Arthoscopic evaluation of the articular cartilage
The distribution of the Outerbridge cartilage scores of the eval-
uated elbows were: Grade I (n = 9), Grade II (n = 26), Grade III
(n = 6) and Grade IV (n = 12).
Cartilage thickness
The mean articular cartilage thickness of the control group
(372.4 lm ± 23.4 lm) was significantly less than samples from
IEWG Grades 2 (858 lm ± 109 lm) and 3 (1075 lm ± 193 lm),
and the mean articular thickness of samples from dogs with IEWG
Grade 0 (655.4 lm ± 58.8 lm) was significantly less than samples
from IEWG Grade 3 (Fig. 1A). The mean articular thickness of sam-
ples from IEWG Grade 1 (684.6 lm ± 87.7 lm) were no different
98 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103

Table 1
Histopathological synovial grading system with the three features evaluated and scored accordingly (Krenn et al., 2006).

Enlargement of the synovial lining cell layer

0 Points One layer of lining cells
1 Point 2–3 Layers of lining cells
2 Points 4–5 Layers of lining cells with a few multinucleated cells which might occur
3 Points More than five layers of lining cells, the lining might be ulcerated and multinucleated cells might occur
Density of resident cells
0 Points Normal cellularity in the synovial stroma
1 Point Slightly increased cellularity in the stroma
2 Points Moderately increased cellularity in the stroma, multinucleated cells might occur
3 Points Greatly increased cellularity, multinucleated giant cells, pannus formation and rheumatoid granulomas might occur
Inflammatory infiltrate
0 Points No inflammatory infiltrate
1 Point Few mostly perivascularly situated lymphocytes or plasma cells
2 Points Numerous lymphocytes or plasma cells, sometimes forming follicle-like aggregates
3 Points Dense band-like inflammatory infiltrate or numerous large follicle-like aggregates

from other samples. The mean articular thickness of the samples in
the control group, Outerbridge Grades 1 (748 lm ± 217 lm), 2
(730.5 lm ± 48.8 lm) and 3 (628 lm ± 182 lm) was significantly
less than samples with Outerbridge Grade 4 (1265 lm ± 182 lm).
A moderate positive correlation was observed between the articu-
lar cartilage thickness, and both the IEWG score (q = 0.37) and the
modified Outerbridge score (q = 0.34).
Mankin score
The mean Mankin score of the control group (0 ± 0) was signif-
icantly lower than the mean Mankin score of samples with IEWG
Grade 1 (7.8 ± 0.7), Grade 2 (8.7 ± 0.8), Grade 3 (11.1 ± 0.26)
(Fig. 1B). The mean Mankin scores for samples with IEWG Grade
0 and Grade 2 were significantly less than Grade 3. The mean Man-
kin score for the control samples was significantly less than the
samples with Outerbridge scores 1 (8.0 ± 1.45), 2 (7.6 ± 0.5), 3
(9.7 ± 0.6) and 4 (10.9 ± 0.4). The mean Mankin score for samples
with an Outerbridge score 2 was significantly less than those with
score 4. A moderate positive correlation was observed between the
Mankin score and both the IEWG score (q = 0.44) and the Outer-
bridge score (q = 0.45).
Synovial score
The distribution of the synovial scores in the FMCP cases was 0
points (n = 3), 1 point (n = 10), 2 points (n = 16), 3 points (n = 6), 4
points (n = 5). The mean synovial score of the control group
(0.8 ± 0.1) was significantly less than for IEWG Grade 3
(2.6 ± 0.4), but did not differ from samples with IEWG Grade 0
(1.6 ± 0.2), 1 (1.5 ± 0.5) or 2 (1.8 ± 0.4) (Fig. 1C). The mean synovial
scores of the control group and samples with an Outerbridge score
1 (0.6 ± 0.4) were less than for samples with score 3 (2.2 ± 0.5) or 4
(2.6 ± 0.4). The mean synovial score of samples with an Outer-
bridge score of 2 (1.8 ± 0.2) was not significantly different from
other samples. A moderate positive correlation was observed be-
tween the synovial score and both the IEWG score (q = 0.45) and
the Outerbridge score (q = 0.30).
Osteocyte density
IEWG Grade 3 (508 cells/mm2 ± 173 cells/mm2) but not IEWG
Grade 2 (699 cells/mm2 ± 277 cells/mm2) (Fig. 1D).
The mean osteocyte density of samples in the control group was
significantly lower than samples with Outerbridge Grade 2
(714 cells/mm2 ± 206 cells/mm2), but not samples with Outer-
bridge Grade 1 (746 cells/mm2 ± 315 cells/mm2), Grade 3
(712 cells/mm2 ± 233 cells/mm2) or Grade 4 (616 cells/
mm2 ± 318 cells/mm2). A moderate negative correlation was ob-
served between the osteocyte density and the IEWG score
(q = 0.33), and a weak negative correlation was observed be-
tween the osteocyte density and the Outerbridge score (q = 0.15).
Bone area
The mean bone area of subchondral bone in control samples
(88.1% ± 2.4%) was significantly greater than the bone area of sub-
chondral bone in dogs with IEWG Grade 1 (73.3% ± 9.4%), Grade 2
(65.1% ± 9.9%) and Grade 3 (73.3% ± 12.7%) (Fig. 1E). The mean
bone area of samples from IEWG Grade 0 (80.6% ± 7.4%) was signif-
icantly greater than samples from IEWG Grade 2. The mean bone
area of the control samples was greater than those with Outer-
bridge score 1 (71.9% ± 11.9%), 2 (75.6 ± 9.9%), 3 (73.2% ± 10.1%)
and 4 (68.5% ± 14.9%). A moderate negative correlation was ob-
served between the bone area and the IEWG score (q = 0.36),
and a weak negative correlation was observed between the bone
area and the Outerbridge score (q = 0.17).
Qualitative assessment
Several disease-related abnormalities became apparent during
the histopathological examination of the affected coronoid pro-
cesses and a number of the sections had ‘cartilage islands’ (13/
53, Figs. 2 and 3) in the subchondral bone. Complete (i.e. full-thick-
ness) fissuring of the subchondral bone of the medial coronoid pro-
cess with intact overlying articular cartilage was identified in nine
cases (Figs. 4 and 5). In one case an incomplete fissure of the sub-
chondral bone of the medial coronoid process with intact overlying
articular cartilage was identified. In all cases there was reduced SO
stain uptake in the articular cartilage directly overlying the fissure
(Figs. 5 and 6).

The mean osteocyte density of samples with IEWG Grade 0 Discussion

(729 cells/mm2 ± 222 cells/mm2) and Grade 1 (842 cells/
mm2 ± 180 cells/mm2) were significantly greater than the osteo-
cyte density of the control group (461 cells/mm2 ± 109 cells/
mm2). The mean osteocyte density of samples with IEWG Grade
1 was also greater than the osteocyte density of samples with
The assessment of existing radiographic (IEWG) and arthro-
scopic (Outerbridge) grading systems in relation to the histopa-
thological changes has not been previously evaluated in naturally
occurring canine joint disease, to our knowledge. Our findings pro-
 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103 99

A. 1600

1400

1200

Cartilage thickness (um)

1000

800

600

400

200

0
1

4
ol

3
T

T
tr

G
U

U
n

W
O

O
Co

IE

IE

IE

IE
B. 12

10
Mean modfiied mankin score

0
1

4
ol

3
T

T
tr

G
U

U
n

W
O

O
Co

IE

IE

IE

IE

C. 3.5

3
Synovial histological score

2.5

1.5

0.5

0
1

4
ol

3
T

G
tr

G
U

U
n

W
O

O
Co

IE

IE

IE

IE

Fig. 1. Comparison of (A) mean articular cartilage thickness (lm ± standard error) between groups (IEWG, Radiographic International Elbow Working Group score; OUT,
Outerbridge score); (B) Mankin score between groups; (C) synovial score between groups; (D) mean osteocyte density (cells/lm2) of subchondral bone between groups; (E)
per cent bone area between groups.

vide clinically valuable information regarding the relevance of Articular cartilage (AC) thickening is reported to be one of the
those grading systems to articular pathology in the MCP. main histopathological changes in joints affected with early OA
100
D. 1200
1000
Osteocyte density (cells/mm2)
800
600
400
200
0
ol
tr
n
Co
E.
(Daubs et al., 2006). In our case series, the magnitude of the AC
thickening reflected the severity of the disease process, as assessed
by arthroscopic or radiographic scoring systems.
A moderate correlation was identified between cartilage pathol-
ogy (Mankin score) in the MCP and disease severity. The severity of
the cartilage changes as assessed by the Mankin score in the sec-
tions of coronoid process evaluated was very high, even in the tis-
sues from the lowest Outerbridge and IEWG graded elbow joints.
The sensitivity of the Outerbridge scoring system to predict the
microscopic changes in articular cartilage of the MCP is weak.
The limitations of the IEWG scoring system become evident be-
cause elbows were scored as radiographically ‘normal’ even when
there were histological lesions in the cartilage. Indeed, the identi-
fication of gross pathological lesions in the radiographically normal
canine elbow has recently been reported (Punke et al., 2009). Mag-
net resonance imaging (MRI) can readily depict alterations in bone
marrow, subchondral bone and cartilage, as well as the surround-
ing soft tissues (Reichle and Snaps, 1999). MRI is regularly used in
human medicine to evaluate joint disease and osteoarthritic
changes (Reichle and Snaps, 1999) but it remains to be seen
whether its sensitivity is sufficient to detect cartilage pathology
in canine MCP disease before radiographic signs have developed.
The histological evaluation of synovial tissue revealed relatively
limited synovial pathology even in cases with marked radiographic
M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103
1
4
3
T
T
G
U
U
W
O
O
IE
IE
IE
IE
Fig 1. (continued)
and arthroscopic evidence of disease. This is similar to the results
of comparable studies in human OA (Krenn et al., 2006), where
high-grade synovitis is primarily associated with immune medi-
ated (rheumatoid) joint disease rather than OA. In our study the
synovial score was most closely correlated to the IEWG grade,
which is not unexpected because osteophytes develop from the
sites of synovial attachment to the joint in the presence of inflam-
mation (Benjamin and McGonagle, 2007). However, it should be
noted that the degree of inflammation in the synovial sample taken
would not necessarily reflect that seen at sites of synovial
attachment.
The osteocyte density was increased in the diseased population
when compared to the control group, with the exception of the
most severely graded samples. This finding was in contrast to pre-
vious reports (Verborgt et al., 2000; Colopy et al., 2004; Danielson
et al., 2006) where a general decrease in osteocyte density has been
reported in all diseased samples no matter what the degree of
severity of cartilage pathology is. On the basis of our results one
could hypothesise that osteocyte density is a measure for different
stages of the disease. An increase in osteocyte numbers may repre-
sent the functional adaptation (remodelling) of bone matrix to an
increased load (Barros et al., 2009) on the MCP. A decreased oste-
ocyte density may represent a response to reduced loading, or
may simply reflect the chronicity of disease and matrix metallo-
 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103
Fig. 2. Photomicrograph of ‘cartilage islands’ () in the subchondral bone of an
affected coronoid process. HE. Bar = 100 lm.
Fig. 3. Photomicrograph of ‘cartilage islands’ () in the same affected coronoid
process as shown in Fig. 2. The cartilage islands are further highlighted in red due to
increased stain uptake. A number of small processing artefacts (clefts) are also
noted (?). Safranin-O. Bar = 100 lm.
Fig. 4. Photomicrograph of intact articular cartilage surface from an affected
coronoid process, with a fissure extending from the osteochondral junction distally
(from  to ). HE. Bar = 100 lm.
101
Fig. 5. Photomicrograph of intact articular cartilage surface from the same affected
coronoid process as in Fig. 4. There is decreased staining intensity of the articular
cartilage directly above the fissure (running from  to ), indicating reduced
proteoglycan content. Safranin-O. Bar = 100 lm.
Fig. 6. Photomicrograph of intact articular cartilage surface showing superficial
fibrillation. There is decreased staining intensity in the cartilage directly above an
incomplete bone fissure, indicating reduced proteoglycan content. Safranin-O.
Bar = 100 lm.
proteinase activity (Clements et al., 2009). Clearly, the comparison
of loading on the FMCP in different ‘grades’ may support or refute
these hypotheses.
The bone area was greatest in the control group and an increase
in bone porosity was observed with increased severity of disease,
consistent with other studies (Danielson et al., 2006). In a previous
report of FMCP bone pathology, the highest percentage porosity
was observed in the area of fragmentation when compared with
other areas of the same sample (Danielson et al., 2006). An increase
in bone porosity may be an indicator for bone remodelling (Ver-
borgt et al., 2000; Colopy et al., 2004). However, whether the
porosity observed represents remodelling of normal bone structure
or a pre-existing difference in bone structure is undetermined.
In several samples the subchondral bone contained discrete foci
of chondrocytes embedded in chondroid matrix, termed ‘cartilage
islands’ (Figs. 2 and 3). Cartilage canals have been reported as a
normal feature in growing Golden Retrievers up to 16 weeks of
age (Wolschrijn et al., 2008). The dogs with cartilage islands in
our population were all older than 24 weeks of age and these
102 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103
may have been retained or developed later as part of the underly-
ing disease process. Retention of cartilage islands may be due to
non-physiological loading and the disturbance of the ossification
process during maturation. Alternatively, the cartilage islands
may be newly formed as part of an attempt by the body to repair
the defect in the subchondral bone after microtrauma.
Subchondral fissures extending into the cartilage were observed
in several samples (Figs. 4 and 5). The uptake of SO stain was re-
duced in the AC directly overlying the fissure compared to the
adjacent AC (Figs. 5 and 6) indicating reduced proteoglycan con-
tent in the cartilage matrix (Jubb and Eggert, 1981). Reduced pro-
teoglycan in these areas suggests that loss of subchondral bone
results in altered cartilage metabolism and that this becomes more
extensive with complete fissuring. Our study tends to support the
hypothesis that primary bone overload with subsequent fissuring
and fragmentation (Danielson et al., 2006) may be the aetiology
of MCP disease rather than primary osteochondrosis – not least
as cartilage (necrotic or otherwise) was not observed in the fissures
in any of these cases (Wolschrijn et al., 2005).
One limitation of our study is that only a single section of the
surgically removed coronoid process of clinical cases was evalu-
ated so there was an underlying assumption that a single section
was sufficiently representative of the entire fragmented MCP. A
second potential limitation is that the control population com-
prised breeds predisposed to FMCP (Labradors) and those not at
high risk of disease (Beagles). The control population was also sig-
nificantly lighter than the diseased cohort. However, a previous
publication reported no differences in the articular cartilage thick-
ness or trabecular porosity between normal dogs from breeds at
low or high risk of MCP disease, so the mix of breeds was unlikely
to be a significant drawback and we decided to consider them to-
gether (Danielson et al., 2006). We observed a wide breed and
weight variation within this study but as no radiographic, arthro-
scopic or histological differences have been previously identified
when correlated with individual weights we included all the sam-
ples for which full information was available.
While it is recognised that different severities of MCP disease
have been reported under the term ‘fragmentation’, these were
not considered as variables that may have contributed to the var-
iability of the data reported. Finally, intra- and inter-observer
agreements were not calculated for the individual tests because
the scoring tests were performed by single experienced individuals
(radiographic score, arthroscopic score, osteocyte density, synovial
score, trabecular porosity) or two individuals together (Mankin
score). The intra-observer agreements which have been reported
for these scoring methods are good to excellent (van der Sluijs
et al., 1992; Brismar et al., 2002; Cameron et al., 2003; Krenn
et al., 2006; Hammond et al., 2008).
Conclusions
Neither the non-invasive radiographic scoring system for FMCP
(IEWG score) nor the minimally invasive arthroscopic scoring sys-
tem (modified Outerbridge score) correlated strongly with the his-
topathological changes in affected tissues evaluated (Mankin and
synovial scores). The severity of cartilage disease in the MCP was
marked, even in cases with minimal radiographic or arthroscopic
pathology of the MCP. More sensitive diagnostic measures of
MCP pathology are required to identify these cases.
Conflict of interest statement
None of the authors of this paper has a financial or personal
relationship with other people or organisations that could inappro-
priately influence or bias the content of the paper.
References
Bardet, J., 1997. Arthroscopy of the elbow in dogs. Part II: the cranial portals in the
diagnosis and treatment of lesions of the coronoid process. Veterinary and
Comparative Orthopaedics and Traumatology 10, 60–66.
Barros, R.R., Degidi, M., Novaes, A.B., Piattelli, A., Shibli, J.A., Iezzi, G., 2009. Osteocyte
density in the peri-implant bone of immediately loaded and submerged dental
implants. Journal of Periodontology 80, 499–504.
Benjamin, M., McGonagle, D., 2007. Histopathologic changes at ‘synovio-entheseal
complexes’ suggesting a novel mechanism for synovitis in osteoarthritis and
spondylarthritis. Arthritis and Rheumatism 56, 3601–3609.
Bobinac, D., Spanjol, J., Zoricic, S., Maric, I., 2003. Changes in articular cartilage and
subchondral bone histomorphometry in osteoarthritic knee joints in humans.
Bone 32, 284–290.
Brismar, B.H., Wredmark, T., Movin, T., Leandersson, J., Svensson, O., 2002. Observer
reliability in the arthroscopic classification of osteoarthritis of the knee. Journal
of Bone and Joint Surgery, British Volume 84, 42–47.
Cameron, M.L., Briggs, K.K., Steadman, J.R., 2003. Reproducibility and reliability of
the Outerbridge classification for grading chondral lesions of the knee
arthroscopically. American Journal of Sports Medicine 31, 83–86.
Carpenter, L.G., Schwarz, P.D., Lowry, J.E., Park, R.D., Steyn, P.F., 1993. Comparison of
radiologic imaging techniques for diagnosis of fragmented medial coronoid
process of the cubital joint in dogs. Journal of the American Veterinary Medical
Association 203, 78–83.
Clements, D.N., Fitzpatrick, N., Carter, S.D., Day, P.J., 2009. Cartilage gene expression
correlates with radiographic severity of canine elbow osteoarthritis. The
Veterinary Journal 179, 211–218.
Colopy, S.A., Benz-Dean, J., Barrett, J.G., Sample, S.J., Lu, Y., Danova, N.A., Kalscheur,
V.L., Vanderby Jr., R., Markel, M.D., Muir, P., 2004. Response of the osteocyte
syncytium adjacent to and distant from linear microcracks during adaptation to
cyclic fatigue loading. Bone 35, 881–891.
Danielson, K.C., Fitzpatrick, N., Muir, P., Manley, P.A., 2006. Histomorphometry of
fragmented medial coronoid process in dogs: a comparison of affected and
normal coronoid processes. Veterinary Surgery 35, 501–509.
Daubs, B.M., Markel, M.D., Manley, P.A., 2006. Histomorphometric analysis of
articular cartilage, zone of calcified cartilage, and subchondral bone plate in
femoral heads from clinically normal dogs and dogs with moderate or severe
osteoarthritis. American Journal of Veterinary Research 67, 1719–1724.
Dingle, J.T., 1981. Catabolin – a cartilage catabolic factor from synovium. Clinical
Orthopaedics and Related Research 156, 219–231.
Ekman, S., Carlson, C.S., 1998. The pathophysiology of osteochondrosis. Veterinary
Clinics of North America Small Animal Practice 28, 17–32.
Gemmill, T.J., Clements, D.N., 2007. Fragmented coronoid process in the dog: is
there a role for incongruency? Journal of Small Animal Practice 48, 361–
368.
Gilbertson, E.M., 1975. Development of periarticular osteophytes in experimentally
induced osteoarthritis in the dog. A study using microradiographic,
microangiographic, and fluorescent bone-labelling techniques. Annals of the
Rheumatic Diseases 34, 12–25.
Grondalen, J., Grondalen, T., 1981. Arthrosis in the elbow joint of young rapidly
growing dogs. V. A pathoanatomical investigation. Nordisk Veterinaermedicin
33, 1–16.
Guthrie, S., Plummer, J.M., Vaughan, L.C., 1992. Aetiopathogenesis of canine elbow
osteochondrosis: a study of loose fragments removed at arthrotomy. Research
in Veterinary Science 52, 284–291.
Hammond, G., Gemmill, T., Mellor, D., Sullivan, M., 2008. Assessment of low-cost
teleradiology for grading elbow dysplasia. Veterinary Journal of Radiology and
Ultrasound 49, 20–25.
Haudiquet, P.H., Rochereau, P.H., 2007. A specific oblique radiography, and
treatment, of a medial coronoid process fracture. Veterinary and Comparative
Orthopaedics and Traumatology 20, 331–334.
Hill, C.L., Seo, G.S., Gale, D., Totterman, S., Gale, M.E., Felson, D.T., 2005. Cruciate
ligament integrity in osteoarthritis of the knee. Arthritis and Rheumatism 52,
794–799.
Jubb, R.W., Eggert, F.M., 1981. Staining of demineralized cartilage. II. Quantitation of
articular cartilage proteoglycan after fixation and rapid demineralization.
Histochemistry 73, 391–396.
Krenn, V., Morawietz, L., Burmester, G.R., Kinne, R.W., Mueller-Ladner, U., Muller, B.,
Haupl, T., 2006. Synovitis score: discrimination between chronic low-grade and
high-grade synovitis. Histopathology 49, 358–364.
Lang, J., Busato, A., Baumgartner, D., Fluckiger, M., Weber, U.T., 1998. Comparison of
two classification protocols in the evaluation of elbow dysplasia in the dog.
Journal of Small Animal Practice 39, 169–174.
Mankin, H.J., Dorfman, H., Lippiello, L., Zarins, A., 1971. Biochemical and metabolic
abnormalities in articular cartilage from osteo-arthritic human hips. II.
Correlation of morphology with biochemical and metabolic data. Journal of
Bone and Joint Surgery, American Volume 53, 523–537.
Mason, T.A., Lavelle, R.B., Skipper, S.C., Wrigley, W.R., 1980. Osteochondrosis of the
elbow joint in young dogs. Journal of Small Animal Practice 21, 641–656.
Meyer-Lindenberg, A., Langhann, A., Fehr, M., Nolte, I., 2002. Prevalence of
fragmented medial coronoid process of the ulna in lame adult dogs.
Veterinary Record 151, 230–234.
Moores, A.P., Benigni, L., Lamb, C.R., 2008. Computed tomography versus
arthroscopy for detection of canine elbow dysplasia lesions. Veterinary
Surgery 37, 390–398.
 M.A. Goldhammer et al. / The Veterinary Journal 186 (2010) 96–103
Ostergaard, K., Andersen, C.B., Petersen, J., Bendtzen, K., Salter, D.M., 1999. Validity
of histopathological grading of articular cartilage from osteoarthritic knee
joints. Annals of the Rheumatic Diseases 58, 208–213.
Ostergaard, K., Petersen, J., Andersen, C.B., Bendtzen, K., Salter, D.M., 1997.
Histologic/histochemical grading system for osteoarthritic articular cartilage:
reproducibility and validity. Arthritis and Rheumatism 40, 1766–1771.
Punke, J.P., Hulse, D.A., Kerwin, S.C., Peycke, L.E., Budsberg, S.C., 2009. Arthroscopic
documentation of elbow cartilage pathology in dogs with clinical lameness
without changes on standard radiographic projections. Veterinary Surgery 38,
209–212.
Reichle, J.K., Snaps, F., 1999. The elbow. Clinical Techniques in Small Animal Practice
14, 177–186.
Rogers, J., Shepstone, L., Dieppe, P., 2004. Is osteoarthritis a systemic disorder of
bone? Arthritis and Rheumatism 50, 452–457.
Samoy, Y., van Ryssen, B., Gielen, I., Walschot, N., van Bree, H., 2006. Review of the
literature: elbow incongruity in the dog. Veterinary and Comparative
Orthopaedics and Traumatology 19, 1–8.
Schulz, K.S., 2003. What’s new in elbow arthroscopy. In: Proceedings of the 13th
Annual American College of Veterinary Surgeons Symposium, Washington DC,
Oktober, pp. 329–333.
Stoker, A.M., Cook, J.L., Kuroki, K., Fox, D.B., 2006. Site-specific analysis of gene
expression in early osteoarthritis using the Pond-Nuki model in dogs. Journal of
Orthopaedic Surgery and Research 1, 8.
Ushiyama, T., Chano, T., Inoue, K., Matsusue, Y., 2003. Cytokine production in the
infrapatellar fat pad: another source of cytokines in knee synovial fluids. Annals
of the Rheumatic Diseases 62, 108–112.
103
van Bree, H.J., van Ryssen, B., 1998. Diagnostic and surgical arthroscopy in
osteochondrosis lesions. Veterinary Clinics of North America Small Animal
Practice 28, 161–189.
van Ryssen, B., van Bree, H., 1997. Arthroscopic findings in 100 dogs with elbow
lameness. Veterinary Record 140, 360–362.
van Ryssen, B., van Bree, H., Simoens, P., 1993. Elbow arthroscopy in clinically
normal dogs. American Journal of Veterinary Research 54, 191–198.
van der Sluijs, J.A., Geesink, R.G., van der Linden, A.J., Bulstra, S.K., Kuyer, R., Drukker,
J., 1992. The reliability of the Mankin score for osteoarthritis. Journal of
Orthopaedic Research 10, 58–61.
Verborgt, O., Gibson, G.J., Schaffler, M.B., 2000. Loss of osteocyte integrity in
association with microdamage and bone remodeling after fatigue in vivo.
Journal of Bone and Mineral Research 15, 60–67.
Wind, A.P., 1986. Elbow incongruity and developmental elbow diseases in the dog:
Part 1. Journal of the American Animal Hospital Association, 711–724.
Wind, A.P., Packard, M.E., 1986. Elbow incongruity and developmental elbow
diseases in the dog: Part 2. Journal of the American Animal Hospital Association,
725–730.
Wolschrijn, C.F., Gruys, E., van der Wiel, C.W., Weijs, W.A., 2008. Cartilage canals in
the medial coronoid process of young Golden Retrievers. The Veterinary Journal
176, 333–337.
Wolschrijn, C.F., Gruys, E., Weijs, W.A., 2005. Microcomputed tomography and
histology of a fragmented medial coronoid process in a 20-week-old golden
retriever. Veterinary Record 157, 383–386.