L'ostéochondrite des condyles fémoraux est une lésion rare. Cette étude multicentrique analyse 892 cas issus d'une série pédiatrique et adulte. Les lésions anatomiques sont plus évoluées chez l'adulte. L'analyse globale montre que la majorité des enfants a un très bon résultat clinique, mais il y a une forte proportion d'enfants dont les radiographies ne sont pas encore normales. Pour les adultes, il y a une forte proportion de patients qui vont bien cliniquement alors que leur radio n'est pas normale.

L'interprétation de la radio standard est délicate. Nous avons défini 3 classes radiographiques, lacunaire, nodulaire et niche vide. L'IRM permet une analyse plus précise de l'interface os-fragment, du cartilage surfacique, du potentiel de croissance et de la vitalité du fragment.

L'arrêt des sports est le traitement de première intention chez l'enfant. Les perforations transchondrales sont une intervention simple de faible morbidité, amenant la guérison dans 48 % des cas en 6 mois, si le cartilage de croissance est ouvert. La fixation du fragment a été faite dans 43 % des cas avec un clapet cartilagineux. Cette intervention donne des résultats moyens qui se dégradent en fonction de la stabilité du fragment. L'opération de Wagner donne des résultats inférieurs aux perforations. L'ablation du séquestre est une opération peu invasive, mais son devenir à long terme est arthrogène, surtout chez l'adulte. Les greffes en mosaïque donnent à moyen terme de bons résultats. Il y a peu de morbidité surtout si les greffes sont prises au dessus de l'échancrure. La greffe de chondrocytes, est de mise en œuvre difficile. Les résultats à moyen terme sont bons pour de larges lésions. L'ostéotomie est logique en cas d'arthrose débutante

1) Le pronostic est meilleur avant la fermeture de la plaque de croissance.

2) L'ostéochondrite nécessite une exploration complémentaire anatomique et fonctionnelle par l'IRM.

3) Il ne faut pas laisser passer le moment de la perforation transchondrale chez l'enfant.

4) Le vissage seul n'est pas toujours suffisant. Il faut améliorer la trophicité, la revascularisation du fragment.

5) Les greffes mosaïques sont préférables à l'ablation du fragment.

6) Les greffes de chondrocytes seront plus utilisées pour demain.

Osteochondritis rarely involves the femoral condyles. Discovery in this localization raises several questions concerning the nature of the articular cartilage, the potential for spontaneous healing, and, in the event of a free fragment, the outcome after its loss or repair.

This multicentric study included 892 pediatric and adult cases, the cutoff between two series being defined by fusion of the inferior growth plate. We excluded medical or surgical osteochondritis, cases involving the patella, osteochondral fractures, juvenile polyosteochondrosis, adult osteonecrosis, and osteochondritis beginning after the age of 50 years.

Mean age at diagnosis was 16.5 years. Mean age at treatment onset was 22 years. Pain was the predominant symptom. 80% of cases were unilateral and 70% involved the medial condyle. The anatomic lesions were different in adults, showing more advanced degradation. At diagnosis, Bedouelle stages Ia and IIb constituted 80% of the cases observed among children while in adults, 66% were Bedouelle stages IIb to IV. Outcome was very good for the majority of children with Hughston clinical stage 4 while half of the x-rays were Hughston stage 3 and 4. There were thus a large percentage of children with abnormal xrays whose disease history was not yet terminated. In the adult series, the percentages of Hughston 3 and 4 was about the same as clinically. The x-rays were rarely perfectly normal since half of the clinical stage 3 patients were noted in stage 4. An abnormal x-ray with a very good clinical presentation was observed in a very large proportion of patients.

It is difficult to interpret the plain x-ray and identify patients with a potentially unfavorable prognosis. We defined three radiographic classes: defect, nodule and empty notch. The Bedouelle classification uses information from all available explorations, particularly MRI and arthroscopy. Numerous therapeutic methods are used. Interruption of sports activities is the first intention treatment for children. Data in the literature and the findings of this symposium do not demonstrate any beneficial effect of immobilization on healing compared with simple abstention from sports activities.

Transchondral perforation is a simple operation with low morbidity. In 85% of cases, it was used for lesions with an intact joint cartilage considered stable in 96% of cases. Healing was achieved in six months for 48% if the growth plate had not fused.

The fragment was fixed in 43% of the cases with a loose cartilage fragment. Outcome was fair but degraded with the state of the joint cartilage and thus the stability of the fragment. Fixation must stabilize the fragment but not prevent further consolidation via osteogenesis. This is why deep perforations are drilled beyond the ossified area and additional osteochondral grafts are used.

The Wagner operation gives less satisfactory results than more complicated procedures.

Removal of a sequestrum is a simple, minimally invasive procedure with an uneventful postoperative period, but in the long term it favors osteoarthritic degradation, especially when performed in adults.

Mosaic grafts give good mid term results. Morbidity is low especially if the grafts are harvested above the notch. The question of chondrolysis around the grafts was beyond the scope of this study.

Chondrocyte grafting is difficult to accomplish and is expensive. The mid term results are good for large lesions.

Osteotomy is logical only in the event of early stage osteoarthritic degradation.

If the plain x-ray reveals a defect (class I), simple interruption of sports activities should be proposed. Two situations can then develop. First, in a certain number of patients, the pain disappears as the defective zone ossifies progressively. Complete cure is frequent before the age of 12 years. In the second situation, the knee remains painful and the x-ray does not change or worsens to a class II nodular formation. In this case an MRI must be obtained to determine whether the joint cartilage is normal. There are two possibilities. First, the osteochondral fragment is viable and most probably will become completely re-integrated, particularly if the lesion is far from the growth plate. Necrosis is the other possibility. Transchondral perforations are needed in this case. If on the contrary the cartilage is altered, there is little hope for spontaneous cure. Arthroscopy may be needed to complete the exploration. Fragments, especially if there is a large surface area, must be fixed. Perforations to favor revascularization are certainly useful here.

In the last situation (class III), the fragment wobbles on a thin attachment or has already fallen into the joint space. This is the type of problem generally observed in adults.

The decision algorithm in adults is the same as in children for the rare nodular aspects (class II). There could be a discussion between transcartilage perforation and fixation. If there are a large number of fragments, fixation may not be fully successful and the lesion might be considered class III.

For class III lesions, three operations can be used: removal of the sequestrum, mosaic bone-cartilage grafts, or autologous chondrocyte grafts.

At the same follow-up, mosaic grafts give better results than excision of sequestra. It may be useful to remove sequestra in a limited number of situations: if there is just a small area of osteochondritis, the lesion is old and partially healed, or the zone is non weight-bearing. For other lesions, we favor mosaic grafts. We still do not have enough follow-up to assess the long-term outcome with these mosaic grafts, but simple excision clearly favors osteoarthritic degradation.

Can chondrocytes grafts be compared with mosaic grafts? Chondrocyte grafts have been used for very large lesions and have given results similar to mosaic grafts. It might also be possible to combine fixation of a loose fragment and a mosaic graft.

1) The prognosis of osteochondritis is better before than after fusion of the growth plate but the lesion does not always heal in children.

2) Presence of osteochondritis requires complementary anatomic and functional exploration to determine the stability and the vitality of the fragment.

3) Attention must be taken to perform transchondral perforations early enough, particularly in children.

4) Screw fixation is not always sufficient. The trophicity of the fragment and its blood supply must be improved.

5) Mosaic grafts are preferable to excision of the fragment.

6) Chondrocyte grafts will be more widely used in the future.