Osteochondritis dissecans-like lesion of the intercondylar fossa of the femur in a dog

Summary An 11-month-old neutered female crossbreed dog was admitted with left pelvic limb lameness. Radiographs and computed tomography identified an osteochondritis dissecans- (OCD) like lesion in the intercondylar fossa of the femur originating from the axial aspect of the medial femoral condyle. Stifle arthrotomy was performed in order to remove the bony fragment. It was concluded that the clinical features and location of this lesion indicate an osteochondritis dessicans-like lesion. Fragment removal was associated with an excellent outcome.


Introduction
Intraarticular osteochondral fragments may be due to osteochondritis dissecans (OCD), due to osteochondrosis, or fracture. The purpose of this article is to report clinical, imaging and pathological features of a lesion of the axial aspect of the medial femoral condyle, resulting in a large osteochondral fragment within the intercondylar fossa of the femur. We propose that this lesion may be an unusual form of stifle OCD. OCD of the canine stifle most commonly affects the articular surface of the lateral femoral condyle, and, less commonly, the articular surface of the medial femoral condyle.

Case presentation
Case history An 11-month-old neutered female crossbreed dog weighing 15 kg was referred to the Queen Mother Hospital for Small Animals for investigation of an intermittent left pelvic limb lameness of approximately two months duration. The lameness became worse after exercise and there was stiffness after rest. There had been an initial response to carprofen a (2 mg/kg daily PO). A traumatic episode had not been noted by the owners prior to the onset of lameness.
The referring veterinarian's radiographs demonstrated a stifle joint effusion and at least two ovoid mineralised fragments superimposed over the medial femoral condyle (Figs. 1 and 2). Upon admission, the dog was 2/5 ths lame on the left pelvic limb (1). There was mild left pelvic limb muscle atrophy, moderate left stifle effusion and slight discomfort on manipulation of the left stifle. Stifle instability was not present.
Computed tomography (CT) of the left stifle revealed a 3 mm by 8 mm, irregularly shaped, bone opacity in the medial half of the intercondylar fossa, associated with a number of smaller mineral opacities caudal to it (Fig. 3). The axial portion of the medial femoral condyle adjacent to the opacities contained an irregular defect which resulted in an approximately 25% reduction of the medial femoral condyle width. Mild subchondral sclerosis was present surrounding this lesion. Synovial fluid from the left stifle had a cell count of 1.3 x 10 9 /l (100% mononuclear cells) and 17.2 g/l protein.

Surgical technique
The dog was pre-medicated with 0.015 mg/ kg iv medetomidine b and 0.3 mg/kg iv morphine c . General anaesthesia was induced with 4 mg/kg iv propofol d and maintained with isoflurane e . Additional analgesia was provided with preoperative carprofen (2 mg/kg iv.) and intraoperative fentanyl f (60 µg iv).
A lateral parapatellar arthrotomy revealed a large ovoid osteochondral fragment within the intercondylar fossa of the left stifle, which obscured the view of the cruciate ligaments (Fig. 4). The fragment was adherent to both cruciate ligaments and was removed by a combination of blunt and sharp dissection (Fig. 5). Smaller mineralised fragments caudal to the main fragment were also removed. The removal of the fragments revealed a large cartilage-covered defect on the axial surface of the medial femoral condyle (Fig. 6). Both cruciate ligaments had frayed fibres where the fragment had been attached, but were otherwise intact and the stifle remained stable. There was no other damage to intraarticular structures. Postoperative analgesia was provided with morphine (0.3 mg/kg iv) every four hours for 24 hours, and 4 mg/kg PO carprofen every 24 hours for seven days.
Postoperatively, strict rest was implemented for two weeks, after which time 10 minute leash walks, were allowed three to four times daily for a further six weeks, followed by a gradual increase in exercise. Approximately six weeks after surgery the dog attended weekly hydrotherapy sessions for four weeks. Twelve weeks after surgery the dog was re-examined at the Queen Mother Hospital for Small Animals. Lameness could not be detected although slight discomfort was elicited at the limit of left stifle extension and mild upper limb muscle atrophy was still evident. Two years after surgery the dog was reported to be sound by the owner.
The avoid mineralised fragment removed from the intercondylar fossa was fixed in 10% neutral buffered formalin, decalcified and processed by routine methods. Sections were stained with haematoxylin and eosin (HE) and a Masson's Trichome stain. The fragment was composed of lamellar bone covered by hyaline cartilage. The subchondral bone was compact; the remaining bone was cancellous. The cartilage had segmental areas of thinning, degeneration and regeneration (Fig. 7). Focal areas of cartilage degeneration were characterised by the loss of the columnar arrangement of chondrocytes, pale eosinophilic staining of the matrix and exposing (unmasking) of the collagen fibres. Large clusters of chondrocytes showing regenerative hyperplasia (chondrone formation) were present

Discussion
Osteochondrosis is an abnormality of endochondral ossification that may affect the epiphyseal, metaphyseal, or apophyseal regions (2). The classic pathological description of osteochondrosis that affects the articular-epiphyseal complex involves focal abnormal thickening of the articular carti-lage. If clefts form in this thickened cartilage an unstable cartilaginous flap may develop, a lesion known as osteochondritis dissecans (OCD) (3). However, in adolescent children, clinical OCD lesions differ from this classic description (4,5). In children, OCD is commonly associated with fracture of an intraarticular necrotic osteochondral fragment which can then receive nutrition within the synovial compartment and undergo a reparative response. With time the necrotic bone undergoes creeping substitution and is replaced with viable bone. The articular cartilage proliferates to cover the entire surface of the loose body and the defect becomes covered in fibrocartilage (4). Osteochondral fragments are also recognised as clinical features of variants of OCD seen in companion animals, affecting, for example, the canine (6, 7) and equine (8,9) hock.
Traumatic fracture is a differential for an intraarticular osteochondral fragment, although obviously a history of trauma would be expected. Since any intraarticular osteochondral fragment can continue to receive nutrition after its separation from the underlying bone, and can therefore undergo remodelling in response to its local environment (10), histopathology would not be ex-pected to differentiate chronic osteochondral fragmentation due to traumatic fracture from that due to OCD. In the case reported herein this was certainly the case. However, the absence of a traumatic episode from the history and the location of the fragment lead us to the hypothesis that it was a developmental lesion rather than a traumatic fracture. Certainly, it is hard to explain how an external trauma could result in a condylar fracture in this location in the absence of other injuries, since the part of the medial condyle that was affected is largely nonweight-bearing and would have been protected from external trauma by virtue of its location deep within the stifle joint. Based on the assumption that the lesion reported herein was developmental, it bears many similarities with OCD as seen in children and in those cases of canine and equine talar OCD associated with osteochondral fragmentation.
OCD has been infrequently reported in the canine stifle (11)(12)(13)(14)(15)(16). More than 90% of lesions affect the articular surface of the lateral condyle (15). This is in contrast with humans where the knee is the most commonly affected joint and the medial femoral condyle is often involved (4,17). OCD lesions bordering the intercondylar fossa of the canine femur have not previously been reported. Cross-sectional imaging proved useful when characterising this lesion prior to surgery, thereby allowing an appropriate surgical approach to be performed.
Recommendations for the treatment of OCD involve flap/fragment removal and the curettage of underlying subchondral bone in order to access the vasculature of the subchondral bone and thereby promote fibrocartilage production (13). This was not necessary here as fibrocartilage had already formed in the defect on the axial aspect of the medial femoral condyle. The spontaneous formation of the fibrocartilage may have been enabled by the non-weight-bearing location of the lesion in this case.
In conclusion, the hypothesis was drawn that the lesion described is an unusual form of stifle OCD. Despite the large size of the fragment and its close association with intraarticular structures, an excellent outcome followed its removal.