Veterinary Surgery

38:173–184, 2009

Treatment of Incomplete Ossification of the Humeral Condyle with

Autogenous Bone Grafting Techniques

NOEL FITZPATRICK, MVB, CertVR, CertSAO,THOMAS J. SMITH, MA VetMB CertSAS, JERRY O’RIORDAN, MVB CertSAS,
and RUSSELL YEADON, MA VetMB

Objective—To report clinical experience with autogenous bone grafting, with and without metallic
implants, for treatment of lameness attributed to incomplete ossification of the humeral condyle
(IOHC).
Study Design—Case series.
Animals—Dogs (n ¼ 8; 9 elbows) with IOHC.
Methods—A transcondylar humeral bone core was removed and the resultant socket grafted using
autogenous bone harvested as either free cancellous bone or a corticocancellous dowel using an
osteochondral autograph transfer system. In 8 elbows, additional support for the humeral condyle
was provided with metallic implants. Postoperative outcome was assessed by clinical, radiographic,
computed tomographic (CT) and owner questionnaire examinations in the short and medium term.
Results—Eight dogs (9 elbows) were treated surgically for IOHC. Graft types were free cancellous
graft (n ¼ 2) or corticocancellous dowel (7). Condylar augmentation was performed using epic-
ondylar cross pins (1); transcondylar AcutrakTM (AT) screw and epicondylar cortical screw (1); and
a single transcondylar AT screw (7). Lameness resolved in 1–12 weeks. Bone bridging was doc-
umented in 7 of 8 elbows assessed by CT examination. Owner questionnaires (6 dogs) assessing
daily functions were available for 7 of 9 elbows (follow-up, 6–45 months). Relevant follow-up
function scores were significantly improved compared with preoperative values. One dog was in-
termittently lame and was administered nonsteroidal antiinflammatory medication.
Conclusion—Use of autogenous bone grafting techniques leads to resolution of lameness attributed
to IOHC. Augmentation of grafts with implants like the AT screw is recommended.
Clinical Relevance—Autogenous bone grafting techniques represent a viable alternative or adjunct
to existing techniques for clinical management of IOHC in the dog.
r Copyright 2009 by The American College of Veterinary Surgeons

INTRODUCTION condyle in Spaniels was reported by Meutstege2 in 1989

T HE HUMERAL condyle in the normal developing
dog has 2 (medial and lateral) centers of ossification,
which are separated by a cartilaginous intermediate zone,
and appear at a mean (  SD) of 14  8 days after birth.1
These ossification centers are reported to unite by 70  14
days of age1 with completion of ossification by 32 weeks
of age. A condition affecting the integrity of the humeral
as a rare finding and subsequently to primarily affect
medium and large breed dogs.3–10 Now commonly
termed incomplete ossification of the humeral condyle
(IOHC), it is overrepresented in the Cocker Spaniel, in
which a recessive mode of inheritance as been proposed.7
IOHC may be a subclinical condition,6,8 and has been
reported as an uncommon cause of forelimb lameness in
dogs.6 Lameness may be mild and intermittent to non-

From Fitzpatrick Referrals, Halfway Lane, Eashing, Surrey, UK.

Presented in part at The British Small Animal Veterinary Association Annual Conference, Birmingham, UK, 2006; European Society
of Veterinary Orthopaedic Traumatology Conference, Munich, Germany, 2006; British Veterinary Orthopaedic Association Autumn
Meeting, Chester, UK, 2006.
Address reprint requests to Noel Fitzpatrick, MVB, CertVR, CertSAO, Fitzpatrick Referrals, Halfway Lane, Eashing, Surrey GU7
2QQ, UK. E-mail: noelF@fitzpatrickreferrals.co.uk.
Submitted March 2008; Accepted July 2008
r Copyright 2009 by The American College of Veterinary Surgeons

0161-3499/09
doi:10.1111/j.1532-950X.2008.00485.x
173
174 BONE GRAFTING TREATMENT OF IOHC
weight bearing in nature4,8,9 and may precede complete
humeral condylar fracture (HCF). It is proposed that
IOHC decreases stability of the humeral condyle predis-
posing to complete fracture often after a minimally trau-
matic event.
IOHC was first diagnosed radiographically as a linear
sagittal radiolucency in the humeral condyle in the region
of the developmental cartilage zone separating the 2 con-
dylar centers of ossification.2 Subsequently bone scinti-
graphy3 and arthroscopic examination8 were reported
useful in establishing a diagnosis but it is now appreciated
that diagnosis may prove elusive using these modalities.
Magnetic resonance imaging or computed tomography
(CT) may be necessary for definitive diagnosis.
Management of IOHC remains controversial. Conser-
vative treatment of IOHC is associated with high rates of
HCF. Marcellin-Little11 reported that 3/7 condyles (43%)
with a partial radiolucent line and 1/12 condyles (8%)
with a complete radiolucent line fractured 11 days to 18
months after diagnosis. Surgical treatment aims to pre-
vent HCF, to encourage osseous fusion of the transcon-
dylar fissure and to resolve lameness in the long term.
Current surgical treatment of IOHC includes use of a
transcondylar screw, either fully or partially threaded,
applied either as a position screw or in lag fashion.3,5–7
Screw placement may be combined with transcondylar
bone tunnels created by drilling (forage) to allow vascular
in-growth.3,6 A single case report suggested use of a
transcondylar cancellous bone graft but this was not per-
formed because of concerns about potential for the bone
graft to communicate with the joint via the transcondylar
fissure.7
After transcondylar screw application, resolution of
lameness usually occurs4; however, complications include
failure to achieve bone union, recurrence of lameness,
fissure widening, loss of transcondylar compression, im-
plant failure, and HCF.3–6,8 Failure to achieve bone
union and condylar stability may result in cyclic loading
of the screw with bending, stress fatigue, and failure.3–6,8
Optimal management of IOHC remains controversial.
Histologic features of the fissure site were consistent with
atrophic nonunion fracture in an English Pointer4 and
were composed of fibrous tissue in 2 Cocker Spaniels with
no evidence of chondrocytes or cartilage matrix.6 These
findings may suggest that IOHC might be approached
similarly to treatment of atrophic nonunion fractures, so
treatment modalities promoting transcondylar bone os-
seous union are worthy of consideration. Autogenous
cancellous bone graft application in the area of incom-
plete ossification has been proposed to optimize bone
formation and remodeling10 by providing of trabeculae
necessary for bone conduction and osteoprogenitor cells,
as well as cytokines and growth factors for osteoinduc-
tion and osteogenesis. No study has reported the effect of
IOHC on condylar stability but resultant instability may
be in part or wholly responsible for observed lameness. In
such cases, use of bone graft alone may not promote bone
healing because of the effects of excessive movement at
the fissure site inhibiting healing, similar to unstable
atrophic nonunions where AO principles support the
combination of a graft with rigid internal fixation to
promote bony union.
The Acutrakt bone screw (AT screw, AcutrakTM,
Acumed, Beaverton, OR) used since 1992 in human pa-
tients,12 is composed of titanium alloy (ASTM F136),
and is a cannulated, headless, tapered, variably-pitched,
self tapping, and fully threaded compression screw. The
AT screw is inserted using a customized cannulated ap-
plication system. The osteochondral autograft transfer
system (OATSt, Arthrex, Naples, FL) has been used in
humans for treatment of articular cartilage defects in-
cluding osteochondritis dissecans. We reasoned that both
systems had features that might facilitate graft collection
and treatment of IOHC.
We hypothesized that removal of a transcondylar core
and bone grafting of the resultant socket combined with
the use of an AT screw would result in bony union across
the condylar fissure and would be consistently associated
with clinical improvement and resolution of lameness.
Our objective is to describe use of autogenous bone
grafting procedures combined with AT screw insertion
for the treatment of lameness attributed to IOHC and to
report the short and medium term clinical outcome in 8
dogs (9 elbows) treated between January 2004 and June
2007.
MATERIALS AND METHODS
Study Design
Medical records (January 2004–July 2007) of dogs that had
surgical treatment of IOHC using transcondylar bone grafting
procedures were identified and included in this study. Elbows
treated for complete HCF where IOHC was suspected as the
underlying cause were excluded. Written consent for IOHC
treatment by transcondylar bone grafting procedures tech-
nique had been obtained from owners after discussion of
available treatment options. Details of clinical, radiographic,
arthroscopic, and CT examinations; surgical technique; post-
operative complications; and resolution of lameness were
recorded.
Historical Examination
Owner interview on admission elucidated the type, inten-
sity, and duration of lameness and any known traumatic
events.
 FITZPATRICK ET AL
Clinical Examination
Full clinical, orthopedic, and neurologic examinations were
performed by a single clinician (N.F.). Elbow examination
included assessment of the response to elbow flexion, exten-
sion, and flexion with simultaneous pronation or supination.
Response to direct digital pressure applied to the lateral ep-
icondyle was noted.
Radiographic Assessment
Radiographic examination of both elbows included cranio-
caudal extended, mediolateral flexed, and mediolateral ex-
tended projections. If no sagittal radiolucent line was evident,
craniocaudal views were repeated with the elbow positioned in
varying degrees of obliquity in an attempt to direct the ra-
diographic beam through the suspected fissure (Fig 1A).6
Postoperative orthogonal radiographic views of the recipient
(elbow) and donor (stifle) graft sites were taken. Radiograph-
ically evident osseous union of the condyle was defined as
absence of a fissure or bridging of the original fissure length by
450% with trabecular bone.
CT Examination
Dogs were positioned in sternal recumbency with the fore-
limbs extended for CT examination (Philips Mx 8000 IDT, 16
Slice, Extended Brilliance Workspace, Philips Healthcare,
Best, the Netherlands) of both elbows. Contiguous 0.8 mm
transverse slices were reconstructed in both sagittal and dorsal
planes and the images recorded. IOHC was recognized as
Fig 1. Typical preoperative imaging in elbows with IOHC.
(A) Craniocaudal radiograph showing typical appearance
of linear sagittal lucency commonly identified as the
‘‘fissure plane.’’ (B) Transverse CT image showing radiolucent
‘‘fissure plane.’’ (C) Arthroscopic image demonstrating
full thickness cartilage fissure (running top-right to mid-left
of image).
175
presence of a radiolucent fissure plane at the isthmus of the
condyle (Fig 1B). Osseous union was defined from CT exam-
ination by an absence of a fissure or bridging of the original
fissure length by 450% with trabecular bone.
Arthroscopic Examination
Four dogs had preoperative arthroscopic examination of
both elbow joints by a single clinician (N.F.) using a 2.4 mm
301 arthroscope (Hopkins, Karl Storz, Tuttlingen, Germany)
connected to a video camera and image-recording device.
Dogs were positioned in dorsal recumbency and each forelimb
held separately in simultaneous extension, abduction,
and pronation over a fulcrum placed under the lateral aspect
of the joint. A standard medial elbow portal13 allowed
evaluation of all major structures including the transcondylar
area, anconeal process, and coronoid process. The transcon-
dylar cartilage was surveyed for evidence of cartilage fracture,
fissuring, or lucency suggestive of IOHC (Fig 1C). Condylar
stability was assessed using an arthroscopic probe and joint
manipulation.
Surgical Technique
One surgeon (N.F.) performed all procedures. Dogs re-
mained in dorsal recumbency with the affected limb extended,
abducted, and pronated. A medial approach to the humeral
condyle after Piermattei and Johnson14 was extended to in-
clude limited exposure of both the caudal and cranial aspects
of the articular surface of the humeral condyle, with retraction
provided using Gelpi self-retaining retractors. Special effort
was made to preserve the medial collateral ligament. Con-
comitant fragmentation of the medial aspect of the coronoid
process was treated by removal of a standardized portion of
the medial aspect of the coronoid process (subtotal coronoid
ostectomy; SCO) using an oscillating saw via a mini-medial
arthrotomy between the flexor carpi radialis and pronator
teres muscles caudal to the medial collateral ligament.15
Placement of Transcondylar AT Screw and Bone
Graft. Direct visibility of the cranial and caudal aspects of
the humeral condyle was achieved and the narrowest isthmus
of the articular surface was marked by 2 temporary 1.1 mm
Kirchner (K)-wires placed within the joint contiguous with the
articular surface (Fig 2). A calibrated K-wire (Acumed) was
driven medial to lateral across the humeral condyle at its most
distal extent, using an inverted AT screw placed over the K-
wire as a spacing guide (Fig 2) to ensure that the maximal
diameter of the screw would not encroach on the articular
cartilage at the narrowest part of the isthmus. When the wire
had penetrated the trans cortex, bone depth was measured
using the etch mark on the calibrated K-wire held against the
scale on the corresponding AT depth gauge (Acumed). The
base of the inverted AT screw placed over the K-wire acted as
a spacer allowing an OATSt reamer (OATSt, Arthrex) of
maximal diameter to be centralized proximal to the AT screw
on the medial aspect of the humeral condyle without en-
croaching on the intended screw position. The reamer position
176 BONE GRAFTING TREATMENT OF IOHC

the guide wire in increments of 3–4 mm and intermittently

Fig 2. Intraoperative photograph showing an inverted AT
screw (AT) threaded over the guide K-wire allowing estimation
of optimal core size and positioning. Two 1.1 mm K-wires (a)
have been placed within the elbow joint to assist identification
of the humeral condylar ‘‘isthmus.’’ b ¼ OATS reamer.
was maintained by advancing a guide drill across the condyle
through the cannulated reamer (Fig 3A). Parallelism of the
drill guide/reamer and the wire/screw is desirable, but in some
dogs with limited humeral condylar bone stock it was neces-
sary to drive the screw parallel to the medial humeral joint
surface rather than parallel to the transverse axis of the
condyle. In these dogs, the screw passed obliquely from disto-
medial to proximolateral, craniodistal to the position of
the intended bone core as marked by the guide drill (Fig 3A
and B)
A hole was prepared for the AT screw using the AT in-
sertion system (Acumed). The guide K-wire was advanced
through the trans cortex, soft tissues, and skin and was se-
cured with wire graspers on the lateral aspect of the condyle to
minimize wire movement. The AT drill bit was advanced over
removed to allow removal of bone debris (Fig 3B). External
drill calibrated markings measured against the cis cortex al-
lowed the advancement of the drill tip to within 2–6 mm of the
transcortex.
The reamer was then repositioned on the central guide drill.
The intended socket depth was 75% of condylar width and
was estimated from preoperative radiographs. Calibrated
markings on the external barrel of the reamer allowed socket
depth measurement during reaming (Fig 3C). An AT screw
2 mm shorter than the drill hole depth was threaded over the
guide wire and inserted to finger tightness using the AT screw
driver (Fig 3D).
Core socket depth and alignment were confirmed using a
calibrated alignment rod before cancellous bone dowel col-
lection. When free autogenous cancellous bone was used, it
was collected from the proximal aspect of the ipsilateral hu-
merus through a small fenestration created using a bone cure-
tte. Corticocancellous bone dowels of appropriate length were
collected from the proximal aspect of the tibia or distal femur
using an OATSt core harvesting chisel 1 mm wider than the
transcondylar recipient socket. The core harvester is a cali-
brated, cylindrical cutting chisel with louvered grooves at 4
equidistant points on the circumference. The louvers engage
the bone core when hammer-tapped into the donor site and
the bone dowel was extirpated by a twisting motion axial to
the harvester, or by slight rocking (‘‘toggling’’) whereupon the
louvers engaged the cancellous bone and broke the dowel off
at its base, which was subsequently removed within the chisel.
Graft dowels were trimmed to fit recipient socket length where
necessary. Grafts were transferred to the recipient socket by
packing the free cancellous graft firmly with a tamping rod to
the level of the cis-cortex or placing a dowel as a press fit using
the OATSt system (Fig 3E).
Humeral epicondylar augmentation was performed in 2 el-
bows. A single 2 mm Steinman pin was inserted in both the
medial and lateral epicondyle in a cross-pin fashion in dog 2

Fig 3. (A) guide K-wire and OATS guide drill bit positioned to optimize bone stock of the humeral condyle in preparation for the
AT screw and autogenous bone graft. An inverted AT screw has been threaded over the guide K-wire to allow estimation of optimal
core size and positioning. Two K-wires assist with identification of the humeral condylar ‘‘isthmus’’; (B) drilling with the custom
tapered drill bit in preparation for AT screw placement; (C) drilling using the depth-calibrated OATS reamer. Red dotted lines
demonstrate typical dimension of socket prepared for autogenous bone graft; (D) insertion of AT screw using custom AT screw-
driver to finger-tightness; (E) insertion of corticocancellous bone graft dowel using OATS ‘‘tamping’’ device.
 FITZPATRICK ET AL
Fig 4. Postoperative mediolateral (A) and craniocaudal (B)
radiographic views showing typical AT screw position.
and a single 2 mm cortical position screw was placed across a
fissure within the lateral epicondyle in dog 7.
Immediate postoperative radiographs confirmed appropri-
ate implant and bone core placement (Fig 4).
Postoperative Care
Carprofen (2 mg/kg orally twice daily) was administered
for 1 week. Strict cage confinement with leash exercise for
toilet purposes only was enforced for the first 6 weeks. For
weeks 6–12 increasing levels of lead exercise were permitted
and thereafter a return to normal exercise was permitted.
Outcome Assessment
Postoperative clinical assessments were performed at 2, 4,
and 8 weeks. Radiographic and CT examinations were sched-
uled at 12 weeks. A questionnaire (Appendix A) mailed to all
owners in August 2006 included 100 mm visual analogue
scores allowing owner estimation of their dog’s daily functions
preoperatively and at follow-up. Each assessment was attrib-
uted a numerical score determined by the position of a mark
placed by the owner along the line. A value of 0 mm was
awarded for poor and 100 mm for excellent outcomes.
Statistical Analysis
Descriptive statistics were calculated for: dog age, body
weight, and duration of lameness at time of surgery; diameter
of the transcondylar core socket; time to resolution of lame-
ness; owner estimation of the success of the procedure and
quality of life. Pre- and postoperative owner assessments of
patient functions were compared for statistical difference by
paired t-test with confidence intervals set at 95%.
RESULTS
Eight dogs (9 elbows) treated for IOHC were included.
Dogs were aged 6 months to 7 years (mean, 43 months;
median, 37 months) and weighed 7.7–28.0 kg (mean,
177
16.8 kg). Two were intact males, 4 were castrated males
and 2 were spayed females. Dogs were Spaniels: Cocker
(n ¼ 3), Springer (3) and Cavalier King Charles (2). Clin-
ical signs on admission ranged from mild intermittent to
high intensity weight bearing lameness of 1–52 weeks du-
ration/elbow (mean, 10 weeks; median, 3 weeks; Table 1).
In affected elbows, pain was consistently observed upon
elbow extension, flexion, flexion with simultaneous pro-
nation or supination of the antebrachium, and on direct
digital pressure applied to the lateral epicondyle.
Assessment of IOHC
Diagnostic tests included radiography (9 elbows), ar-
throscopy (4) and CT examination (7; Table 1). Preop-
erative CT examination was not performed in dogs 7 and
8 because the owners chose immediate arthroscopic in-
vestigation during which condylar instability and a de-
finitive transcondylar fissure were observed (Fig 1C).
Bilateral IOHC was evident on referral radiographs for
dog 2, but HCF occurred in the right elbow before ad-
mission for treatment.
IOHC Treatment
An autogenous corticocancellous dowel was used in 7
elbows and free cancellous graft in 2 elbows. Transcon-
dylar bone graft core diameter ranged from 4 to 7 mm
(mean, 5.7 mm). Condylar augmentation was performed
using epicondylar cross pins (1 elbow), transcondylar AT
screw and epicondylar cortical screw (1) and a single
transcondylar AT screw (6; Table 3). Dogs 1 and 4 had
concomitant fragmentation of the medial coronoid pro-
cess treated by SCO.
Outcome
Complications included soft tissue infection at the
donor site in dog 2, and seroma formation at the reci-
pient site in dog 5; both responded to medical treatment.
Time to resolution of lameness ranged from 4 to 84
days (mean, 35 days). Partial ( 50% width of central
portion of condyle) or complete bone union was identi-
fied in 7/9 elbows by CT examination, 11–16 weeks post-
operatively (Figs 5 and 6). Failure of bone union
observed in dog 5 occurred where free cancellous graft
was combined with an AT screw. Dog 3 was lost to CT
follow-up (Table 2).
Owner Assessment
Owner questionnaires were returned for 6/8 dogs (7/9
elbows) with follow-up ranging from 6 to 45 months
postoperatively (mean, 20 months; median, 14 months).
178 BONE GRAFTING TREATMENT OF IOHC

Table 1. Summary Data for 8 Dogs with Incomplete Ossification of the Humeral Condyle
Preoperative
Dog Breed Age (Months) Sex Weight (kg) Limb Lameness (Weeks) Radiography CT Arthroscopy

1 Cocker Spaniel 43 M 20.0 Left 52  þ MCD

Right N/A   FCP
2 Springer Spaniel 69 FN 20.45 Left 3 þ þ 
Right N/A þ  N/A
3 Springer Spaniel 31 MN 21.0 Left 3 þ þ Fissure
Right N/A   
4 Cavalier King Charles Spaniel 21 MN 7.7 Left 1 þ þ FCP
Right N/A   
5 Cocker Spaniel 85 FN 17.8 Left 8 þ þ 
Right 8 þ þ 
6 Cavalier King Charles Spaniel 28 MN 10.0 Left N/A   
Right 6  þ Fissure
7 Cocker Spaniel 4 M 9.5 Left 1 þ N/A Fissure
Right N/A   FCP
8 Springer Spaniel 60 MN 28.0 Left 2 þ N/A Fissure
Right N/A   

MCD, disease of the medial aspect of the coronoid process; FCP, fractured coronoid process; M, male; MN, male neutered; FN, female neutered;
IOHC, incomplete ossification of the humeral condyle; HCF, humeral condylar fracture; þ ¼ positive finding;  ¼ nothing abnormal detected; N/A, not
applicable or not performed; , radiography by referring veterinarian identified IOHC, before HCF and referral.

At that time 6/7 elbows were free from lameness. Though
a questionnaire was not performed for dog 9, the owner
reported at 5 months that the dog was without lameness
and was fully working as a gun dog. Dog 1 was inter-
mittently lame on the operated forelimb and was admin-
istered nonsteroidal antiinflammatory medication when
lameness was observed. Two of 6 dogs were being ad-
ministered nutraceutical supplementation at follow-up.
Owners (n ¼ 6) scored: (1) success of the procedure with
a range from 83 to 100 mm (mean, 93 mm; median, 93 mm);
(2) current quality of life from 68 to 100 mm (mean, 87 mm;
median, 87 mm); and (3) whether they would have the pro-
cedure done again from 83 to 100 mm (mean, 92 mm; me-
dian 95 mm). Ten of 12 assessments of daily functions made
by owners were significantly (paired t-test) improved post-
operatively (Table 3). Only 2 control functions did not show
significant improvement: ‘‘How well can your dog sit down
without pain or hesitation?’’ and ‘‘How well can your dog
rise on hind legs without pain or hesitation?’’
DISCUSSION
Identification of IOHC lesions by radiographic imag-
ing alone may be inadequate. IOHC not detected in 2
dogs by radiographic examination was identified by CT
examination (Table 1). Multiple craniocaudal radio-
graphic views of both elbows, in various degrees of
obliquity, are routinely taken by the authors when there

Fig 5. CT image. 12 weeks postoperative. There is complete Fig 6. CT image. 12 weeks postoperative. There is partial
osseous integrity across the fissure line where graft has been osseous integrity across the fissure line. Note persistence of
placed. Note AT screw in distocentral humeral condyle. CT, fissure line adjacent to articular surface extending  20% of
computed tomographic. width of condyle. CT, computed tomographic.
 FITZPATRICK ET AL 179

Table 2. Surgical Method, Outcome and Follow-Up

Follow-up
Bone Union—Fissure
Core Diameter Transcondylar Ancillary Lameness Resolution Owner
Dog Graft (mm) Implant Implant Complications (Weeks) Radiography CT Questionnaire

1 Dowel 6 — — Infection 7 Present Present Yes

(Donor site)
2 Dowel 7 — Cross-pins Seroma 4 Present Present No
(Recipient
site)
3 Dowel 7 AT screw — — 8 Present Not available Yes
4 Cancellous — AT screw — — 2 Present Present Yes
5 (L) Cancellous — AT screw — — 4 Absent Absent Yes
5 (R) Dowel 4 AT screw — — 11 Present Present Yes
6 Dowel 4 AT screw — — 7 Present Present Yes
7 Dowel 5 AT screw AO-type 2 mm — 1 Present Present Yes
screw
8 Dowel 7 AT screw — — 6 Present Present No

AT, AcutrakTM; CT, Computed Tomography.

is intermittent thoracic limb lameness or if lameness is cion for IOHC even in the absence of a radiographically
localized to the elbow joint but no other potential cause discernable fissure. When IOHC is suspected, CT exam-
of lameness has been identified on radiographic or ar- ination is performed before (or in preference to)
throscopic examination. Breed (i.e., Spaniel), physical arthroscopic examination. We consider that both
findings (pain on deep digital palpation of the lateral ep- arthroscopy and CT examinations should be performed
icondyle, pain on elbow extension) and radiographic to aid diagnosis of disease of the medial aspect of the
findings other than an intercondylar fissure, such as re- coronoid process. Our experience agrees with previous
modeling of the lateral epicondyle, may increase suspi- reports suggesting that CT examination is potentially
more sensitive than radiography for diagnosing
Table 3. Summary of Pre- and Postoperative Visual Analogue Scores IOHC10,16 and we consider it a more appropriate method
(VAS) Assessed by Owners for assessing postoperative bone healing/bridging of the
VAS Mean  SD in mm transcondylar fissure.
(Median, Range) There are many potential donor sites for autogenous
Owner Assessment N Preoperative Postoperative t-Test Value P-Value bone graft collection. Factors influencing site selection in-
Walk 6 14.8  13 77.3  21.2 5.82 0.002
clude the length and diameter of the required dowel, the
(27.5, 0–32) (83.5, 50–100) contour of the cortical bone of the cis cortex (the chisel has
Climb (up) 6 39.3  33.4 86.8  13.0 3.24 0.023 a tendency to slip off curved bone surfaces and resulting
(74.5, 0–79) (96.3, 69–100) dowels have a curved cortical bone surface), and patient
Climb (down) 6 40.5  34.2 87.3  12.3 0.04 0.029 positioning. We currently recommend the distal aspect of
(74.8, 0–80) (96.3, 72–100)
Jump (up) 6 30.0  30.7 80.8  17.7 3.67 0.014
the femur because it provides a suitable-sized graft, it is
(61, 0–79) (95.5, 54–100) technically easier to seat the cutting chisel perpendicular to
Jump (down) 6 35.7  30.1 86.7  12.5 3.54 0.016 the bone when compared with other sites, and it is easily
(61.5, 0–81) (96.0, 71–100) accessible without associated morbidity. We find that the
Exercise tolerance 6 35.8  31.0 89.0  10.5 4.00 0.010 cortical surface of a dowel harvested from the distal fem-
(68.3, 0–78) (96.0, 70–100)
Sit 6 17.7  43.3 90.8  10.6 2.50 0.054
oral condyle provides a suitable flat surface which facil-
(52.5, 0–91) (97, 70–100) itates directional ‘‘tamping’’ into the recipient site. The
Lie 6 38.8  38.5 90.8  11.1 3.15 0.025 proximal aspects of the humerus and tibia are other po-
(84.8, 0–90) (97, 69–100) tential donor sites, but cis cortex perpendicularity with the
Rise (fore limbs) 6 37.0  35.4 90.8  11.5 3.49 0.018 long axis of the cancellous graft is difficult to achieve and
(73.0, 0–91) (97, 68–100)
Rise (hind limbs) 5 55.2  42.2 94.8  3.6 2.04 0.111
may cause problems during graft insertion.
(87, 8–89) (95, 91–100) Condylar instability observed in dog 7 and HCF in dog
Run 5 35.0  37.5 92.2  5.8 3.02 0.039 2 (before admission) indicates that there is a subgroup of
(75.5, 0–78) (97, 84–100) IOHC that may be at increased risk of HCF after initial
Head nod 6 42.2  37.2 90.8  9.8 3.11 0.027 diagnosis. These dogs may be identified subjectively after
(79.5, 0–81) (98.5, 74–100)
radiographic or arthroscopic examination. We strongly
180 BONE GRAFTING TREATMENT OF IOHC
discourage conservative management of such dogs and
propose that initial management should include appro-
priate limb immobilization to stabilize the elbow followed
by surgical management as a priority; an approach sup-
ported by the positive outcome achieved in dog 7.
In this case series, we managed IOHC similar to an
atrophic nonunion by providing rigid support and a bone
graft. Use of autogenous cancellous bone remains the
gold standard for treatment of nonunions in veterinary
orthopedics.17 A corticocancellous bone graft harvested
as a dowel resulted in bone union in all elbows subse-
quently examined with CT (Table 3). The observed fail-
ure of satisfactory bone union in dog 5 was not associated
with evidence of osteomyelitis or implant loosening and
lameness resolution suggests that mechanical stability of
the condyle was achieved. Factors that may explain fail-
ure of bone union include damage to the structural and
cellular elements of the free cancellous graft from exces-
sive packing of the graft resulting in destruction of nor-
mal architecture or/and cellular damage although there is
no objective data to support these hypotheses. However,
as a precaution we no longer use free cancellous graft
preferring instead the biologically undisturbed structure
of a corticocancellous dowel.
Dowels are composed primarily of cancellous bone
and gross observations indicate that this has low me-
chanical strength. For this reason and after experience
with dog 1, we concurrently use a transcondylar bone
screw or other form of condylar augmentation. Dog 1
was especially strictly cage-confined during the immediate
postoperative period and condylar union was docu-
mented by CT, but risk of fracture was considered high
perioperatively. Condylar bone union occurred consis-
tently at the level of the dowel but not distal to the AT
screw or proximal to the dowel on transverse CT recon-
structions. We suggest the mechanical strength of this
intercondylar ‘‘spot-weld’’ may increase with increasing
dowel diameter with dog 1 having a particularly large
dowel to isthmus ratio (6 mm dowel, 9 mm isthmus).
Dowel diameter is intrinsically limited by humeral con-
dylar isthmus dimension and by the concurrent use of a
transcondylar screw.
Use of an AT screw allows placement of a mechan-
ically robust but narrow implant while maximizing bone
dowel diameter. The cannulated system allows accurate
insertion, using the guide wire of the screw and the
reamer centralizer as trajectory guides for the screw and
bone dowel respectively, without need for fluoroscopy,
although fluoroscopic guidance may further facilitate ac-
curate screw placement. Biomechanical assessment of AT
screw use in horses has shown generally comparable to
superior mechanical properties compared with AO can-
cellous and cortical screws12,18–21 and the screws have
been used in Thoroughbred racehorses with encouraging
results.22,23 The AT screw is headless, allowing placement
under the surface of the bone, limiting implant impinge-
ment of soft tissues. Its fully threaded, tapered nature
provides constant new bone purchase as it is inserted,
minimizing strip-out and maximizing pullout strength,
providing strong internal fixation.
From this study, we cannot differentiate the relative
contributions of the bone graft or AT screw to resolution
of lameness. Considering the low number of cases and
variability of patient data (Tables 1 and 2) we can only
speculate that early resolution of lameness is attributable
to condylar stability provided by the AT screw. We can-
not definitively claim that bone grafting contributes to
clinical improvement especially considering dog 5 where
bone union was not apparent with radiographic or CT
assessment. However in dogs 1 and 2, bone union and
resolution of lameness occurred after transcondylar bone
dowel insertion without AT screw use, supporting our
hypothesis that promotion of bone union will lead to
medium to long-term resolution of lameness. Trabecular
‘‘spot-weld’’ was consistently observed in all elbows with
corticocancellous dowel grafts evaluated by CT. In con-
trast to an inert metallic implant which is susceptible to
cyclic fatigue, the dowel should theoretically function as a
biologically active transcondylar bridge capable of re-
sponding to chronic stress by active regeneration, repair,
and remodeling in keeping with Wolff’s law.
Ideally, a transcondylar needle biopsy would have
been performed during follow-up to establish the fate of
the bone grafts but was not undertaken because of con-
cern for morbidity in healthy dogs, free from clinical
signs. Subsequent potential for histologic examination
postmortem will be dependent on owner consent. Despite
this limitation, implant failure has not occurred in any
dog (7/9 elbows with 6–45 months follow-up). The sig-
nificance of this observation for a small case series cannot
be established but given the published estimate of  1 in
5 chance of fracture within 18 months of diagnosis of
IOHC without surgical management11 we attribute out-
come in our dogs, at least in part, to bone grafting.
Dogs 1 and 4 had concomitant fragmentation of the
medial coronoid process treated by SCO. Resolution of
lameness after SCO is typically noted 1–12 weeks post-
operatively15 which is similar to the resolution of lame-
ness we noted (Table 3). It is impossible to definitively
establish the influence of concurrent medial coronoid
disease on the pre- and postoperative clinical, radio-
graphic, and owner assessments with a small case number
but we postulate that concurrent disease may either have
no influence on, or more likely negatively influenced, the
expected clinical outcome (e.g., potentially in dog 1).
Assessment of postoperative outcome in this case series
is subjective. We consider the use of CT imaging to assess
condylar union to have salient advantages over conven-
 FITZPATRICK ET AL
tional radiography, supported by similar opinions regard-
ing use of CT for initial diagnosis of IOHC.10,16 Presence
of metallic implants may affect CT image quality, which
may have influenced our assessment of treatment efficacy.
However, CT resolution was such that the fissure could
still be perceived distal to the AT screw, but could not be
seen throughout the zone of transplant proximal to the
screw in the same plane (where a complete isthmus fissure
had previously been evident), intimating efficacy of the
bone dowel in bridging the fissure. Whereas specific
quantitative data are lacking for the canine humerus, the
material properties of titanium implants (such as the AT
screw) may influence CT image quality to a lesser extent
than steel implants.24 Arthroscopic examination allowed
investigation of concurrent elbow disease and confirma-
tion of IOHC in 1 elbow that was considered radiograph-
ically normal. The sensitivity of imaging modalities used
to diagnose and assess treatment methods for IOHC has
not been reported and represents a limitation to this and
previously reported studies.
To our knowledge, no study has established the
requirements for successful treatment of IOHC. We
consider bone union across the condylar fissure to be the
optimal outcome and considered a value of 50%
healing as a realistic goal based on the knowledge that
the zone that could be bridged by bone was intrinsically
curtailed by the AT screw and its spacing from the
joint surface in most elbows. The significance of 50%
healing of the humeral condyle is subjective but during
short term follow-up was sufficient to contribute to res-
olution of lameness and prevention of HCF in our opin-
ion, within the previously discussed limitations of this
study.
Surgery was performed without benefit of intraoper-
ative fluoroscopy. While not a prerequisite for perform-
ing this technique, fluoroscopic guidance may have
advantages for accurate assessment of bone dowel and
implant trajectories, thus maximizing bone dowel diam-
eter. Placement of guide wires might also be performed
without entering the joint space. However, the technique
used was reproducibly accurate and allows use by sur-
geons without access to fluoroscopy.
We concluded that use of an autogenous corticocan-
cellous bone dowel combined with an AT screw is an
effective and appropriate treatment for lameness attrib-
uted to IOHC. This technique has led to consistent res-
olution of lameness and focal bony condylar union as
assessed by CT. No major complications were noted
during short and medium term follow-up. This technique
requires more specialized instrumentation and refined
technique than previously reported methods but we ad-
vocate consideration of this treatment approach because
of its theoretical benefits and favorable outcome, albeit in
a small cohort of dogs.
181
ACKNOWLEDGMENTS
We wish to thank the primary care clinicians who referred
these dogs; Bernard Walsh for his assistance with the CT
images; Tim Vojt for his help with preparation of Fig
3A–E; and Kei Hayashi for his constructive comments
during preparation of this report.
REFERENCES
1. Hare WC: The ages at which the centers of ossification ap-
pear roentgenographically in the limb bones of the dog.
Am J Vet Res 90:825–835, 1961
2. Meutstege FJ: Incomplete humeral condylar facture in the
canine humerus as a cause of obscure elbow lameness.
Proceedings of the 11th Annual Meeting of the Veterinary
Orthopedic Society, Park City, UT, 1989
3. Butterworth SJ, Innes JF: Incomplete humeral condylar frac-
tures in the dog. J Small Anim Pract 42:394–339, 2001
4. Gundi G, Martini FM, Zannicheli S, et al: Incomplete hu-
meral condylar fracture in two English Pointer dogs. Vet
Comp Orthop Traumatol 18:243–245, 2005
5. Kaderly RE, Lamothe M: Incomplete humeral condylar
fracture due to minor trauma in a mature Cocker Spaniel. J
Am Anim Hosp Assoc 28:361–364, 1992
6. Marcellin-Little DJ, DeYoung DJ, Ferris KK, et al: Incom-
plete ossification of the humeral condyle in Spaniels. Vet
Surg 23:475–487, 1994
7. Marcellin-Little DJ, Roe SC, DeYoung DJ: What is your
diagnosis? J Am Vet Med Assoc 209:727–728, 1996
8. Meyer-Lindenberg A, Heinen V, Fehr M, et al: Incomplete
ossification of the humeral condyle as a cause of lameness
in dogs. Vet Comp Orthop Traumatol 3:187–194, 2002
9. Robin D, Marcellin-Little DJ: Incomplete ossification of the
humeral condyle in two Labrador Retrievers. J Small Anim
Pract 42:231–234, 2001
10. Rovesti GL, Fluckiger M, Margini A, et al: Fragmented co-
ronoid process and incomplete ossification of the humeral
condyle in the Rottweiler. Vet Surg 27:354–357, 1998
11. Marcellin-Little DJ: Incomplete ossification of the humeral
condyle in dogs, in Bonagura JD (ed): Kirk’s Current Vet-
erinary Therapy XIII. Philadelphia, PA, Saunders, 1999,
pp 1000–1004
12. Wheeler DL, McLoughlin SW: Biomechanical assessment of
compression screws. Clin Orthop Rel Res 350:237–245,
1998
13. Beale BS, Hulse DA, Schulz KS, et al: Chapter 4; ar-
throscopically assisted surgery of the elbow joint, in Beale
BS, Hulse DA, Schulz KS, et al: (eds): Small Animal Ar-
throscopy. Philadelphia, PA, Saunders, 2003, pp 51–79
14. Piermattei DL, Johnson KA: Approach to the medial aspect
of the humeral condyle and medial coronoid process of the
ulna by an intermuscular septum, in Piermattei DL, John-
son KA (eds): Atlas of Surgical Approaches to the Bones
and Joints of the Dog and Cat (ed 4). Philadelphia, PA,
Saunders, 2004, pp 226–231
15. Fitzpatrick N Subtotal coronoid ostectomy (SCO) for treat-
ment of medial coronoid disease: A prospective study of
182 BONE GRAFTING TREATMENT OF IOHC
263 dogs (429 elbows) evaluating short and medium term
outcome. Proceedings of the 35th Annual American College
of Veterinary Surgeons Symposium, San Diego, CA, pp
416–420, 2005
16. Carrera I, Hammond GJ, Sullivan M: Computed tomo-
graphic features of incomplete ossification of the canine
humeral condyle. Vet Surg 37:226–231, 2008
17. McLaughlin RM, Roush JK: Autogenous cancellous and cor-
ticocancellous bone grafting. Vet Med 93:1071–1074, 1998
18. Bueno AC, Galuppo LD, Taylor KT, et al: A biomechanical
comparison of headless tapered variable pitch and AO
cortical bone screws for fixation of a simulated slab fracture
in equine third carpal bones. Vet Surg 32:167–177, 2003
19. Eddy AL, Galuppo LD, Stover SM, et al: A biomechanical
comparison of headless tapered variable pitch compression
and AO cortical bone screws for fixation of a simulated
midbody transverse fracture of the proximal sesamoid bone
in horses. Vet Surg 33:253–262, 2004
20. Galuppo LD, Stover SM, Jensen DG, et al: A biomechanical
comparison of headless tapered variable pitch and AO
cortical bone screws for fixation of a simulated lateral
condylar fracture in equine third metacarpal bones. Vet
Surg 30:332–340, 2001
21. Galuppo LD, Stover SM, Jensen DG: A biomechanical com-
parison of equine third metacarpal condylar bone fragment
compression and screw pushout strength between headless
tapered variable pitch and AO cortical bone screws. Vet
Surg 31:201–210, 2002
22. Galuppo LD, Simpson EL, Greenman SL, et al: A clinical
evaluation of a headless, titanium, variable-pitched, ta-
pered, compression screw for repair of non-displaced lat-
eral condylar fractures in thoroughbred racehorses. Vet
Surg 35:423–430, 2006
23. Hirsch JE, Galuppo LD, Graham LE, et al: Clinical evalu-
ation of a titanium, headless variable-pitched tapered can-
nulated compression screw for repair of frontal plane slab
fractures of the third carpal bone in thoroughbred race-
horses. Vet Surg 36:178–184, 2007
24. Jäkel O, Reiss P: The influence of metal artefacts on the range
of ion beams. Phys Med Biol 52:635–644, 2007
FITZPATRICK ET AL 183

APPENDIX A
184 BONE GRAFTING TREATMENT OF IOHC